research paper on experiments on animals

Animal experimentation: A look into ethics, welfare and alternative methods

• November 2017
• Revista da Associação Médica Brasileira 63(11):923-928
• 63(11):923-928

• Universidade Federal de Goiás

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations
• Kardelen Kocaman Kalkan
• Meral Kekeçoğlu

• BMC Med Educ
• Harpreet Kaur and Nidhi Srivastava

• Marie P.F. Corradi

• Gabriela Vitória de Oliveira

• Katsuyoshi Kudo

• INT J MOL SCI

• Ilaria Baldelli

• Susanna Penco

• Caio Vinicius Botelho Brito
• Rosa Helena de Figueiredo Chaves Soares

• INT J IMMUNOPATH PH

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

• Search Menu
• Sign in through your institution
• Advance articles
• Themed Issues
• Author Guidelines
• Open Access
• About ILAR Journal
• About the Institute for Laboratory Animal Research
• Editorial Board
• Advertising and Corporate Services
• Self-Archiving Policy
• Dispatch Dates
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Introduction, experimental design: initial steps, design of the animal experiment, experimental design: final considerations.

• < Previous

Practical Aspects of Experimental Design in Animal Research

Paula D. Johnson, D.V.M., M.S., is Executive Director, Southwest Association for Education in Biomedical Research, University of Arizona, Tucson; David G. Besselsen, D.V.M., Ph.D., is Veterinary Specialist and Chief, Pathology Services, University Animal Care, University of Arizona, Tucson.

• Article contents
• Figures & tables
• Supplementary Data

Paula D. Johnson, David G. Besselsen, Practical Aspects of Experimental Design in Animal Research, ILAR Journal , Volume 43, Issue 4, 2002, Pages 202–206, https://doi.org/10.1093/ilar.43.4.202

• Permissions Icon Permissions

A brief overview is presented of the key steps involved in designing a research animal experiment, with reference to resources that specifically address each topic of discussion in more detail. After an idea for a research project is conceived, a thorough review of the literature and consultation with experts in that field are pursued to refine the problem statement and to assimilate background information that is necessary for the experimental design phase. A null and an alternate hypothesis that address the problem statement are then formulated, and only then is the specific design of the experiment developed. Likely the most critical step in designing animal experiments is the identification of the most appropriate animal model to address the experimental question being asked. Other practical considerations include defining the necessary control groups, randomly assigning animals to control/treatment groups, determining the number of animals needed per group, evaluating the logistics of the actual performance of the animal experiments, and identifying the most appropriate statistical analyses and potential collaborators experienced in the area of study. All of these factors are critical to designing an experiment that will generate scientifically valid and reproducible data, which should be considered the ultimate goal of any scientific investigation.

Experimental design is obviously a critical component of the success of any research project. If all aspects of experimental design are not thoroughly addressed, scientists may reach false conclusions and pursue avenues of research that waste considerable time and resources. It is therefore critical to design scientifically sound experiments and to follow standard laboratory practices while performing these experiments to generate valid reproducible data ( Bennett et al. 1990 ; Diamond 2001 ; Holmberg 1996 ; Larsson 2001 ; Sproull 1995 ; Weber and Skillings 2000 ; Webster 1985 ; Whitcom 2000 ). Data generated by this approach should be of sufficient quality for publication in well-respected peer-reviewed journals, the major form of widespread communication and archiving experimental data in research. This article provides a brief overview of the steps involved in the design of animal experiments and some practical information that should also be considered during this process.

Literature Search

A thorough search of the scientific literature must be performed to determine what is known about the focus of the study. The search should include current and past journal articles and textbooks, as well as information available via the internet. Journal searches can be performed in any number of appropriate journal databases or indexes (e.g., MEDLINE, TOXLINE, PUBMED, NCBI, AGRICOLA). The goals of the literature search are to learn of pertinent studies and methods, identify appropriate animal models, and eliminate unnecessary duplication of research. The “3Rs” of animal research ( Russell and Burch 1959 ) should also be considered at this stage: reduction of animal numbers, refinement of methods, and replacement of animals by viable nonanimal alternatives when these exist. The literature search is also an important component of an institutional animal care and use committee (IACUC 1 ) protocol submission to provide evidence that the project is not duplicative, that alternatives to the use of animals are not available, and that potentially painful procedures are justified.

Scientific Method

The core aspect of experimental design is the scientific method ( Barrow 1991 ; Kuhn 1962 ; Lawson 2002 ; Wilson 1952 ). The scientific method consists of four basic steps: (1) observation and description of a scientific phenomena, (2) formulation of the problem statement and hypothesis, (3) use of the hypothesis to predict the results of new observations, and (4) the performance of methods or procedures to test the hypothesis.

Problem Statement, Objectives, and Hypotheses

It is critical to define the problem statement, objectives, and hypotheses clearly. The problem statement should include the issue that will be addressed experimentally and its significance (e.g., potential application to human or animal health, improved understanding of biological processes). Objectives should be stated in a general description of the overall goals for the proposed experiments and the specific questions being addressed. Hypotheses should include two distinct and clearly defined outcomes for each proposed experiment (e.g., a null and an alternate hypothesis). These outcomes may be thought of as the two experimental answers to the specific question being investigated: The null hypothesis is defined as no difference between experimental groups, and the alternate hypothesis is defined as a real difference between experimental groups. Development of a clearly stated problem statement and the hypotheses are necessary to proceed to the next stage of the experimental design process, although they obviously can (and likely will) be modified as the process continues. Examples of a problem statement and various types of hypotheses follow:

Problem statement: Which diet causes more weight gain in rats: diet A or diet B?

Null hypothesis: Groups are expected to show the same results (e.g., rats on diet A will gain the same amount of weight as rats on diet B).

Alternate hypothesis: Experimental groups are expected to show different results (e.g., rats will gain more weight on diet A than diet B, or vice versa).

Nontestable hypothesis: A result cannot be easily defined or interpreted (e.g., rats on diet A will look better than rats on diet B). What does “better” mean? Its definition must be clearly stated to create a testable hypothesis.

Identification of Animal Model

In choosing the most appropriate animal models for proposed experiments, we offer the following recommendations: (1) Use the lowest animal on the phylogenic scale (in accordance with replacement, one of the 3Rs). (2) Use animals that have the species- and/or strain-specific characteristics desirable or required for the specific study proposed. (3) Consider the costs associated with acquiring and maintaining the animal model during the period of experimentation. (4) Perform a thorough literature search, network with colleagues within the selected field of study, and/or contact commercial vendors or government-supported repositories of animal models to identify a potential source of the animal model. (5) Consult with laboratory animal veterinarians before final determination of the animal model.

Identification of Potential Collaborators

The procedures required to carry out the experiments will determine what, if any, additional expertise is needed. It is important to identify and consult with potential collaborators at the beginning of project development to determine who will be working on the project and in what capacity (e.g., as coinvestigators, consultants, or technical support staff). Collaborator input into the logistics and design of the experiments and proper sample acquisition are critical to ensure the validity of the data generated. Core facilities at larger research institutions provide many services that involve highly technical procedures or require expensive equipment. Identification of existing core facilities can often lead to the development of a list of potential intramural collaborators.

Research Plan

A description of the experimental manipulations required to address the problem statement, objectives, and hypotheses should be carefully devised and documented ( Keppel 1991 ). This description should specify the experimental variables that are to be manipulated, suitable test parameters that accurately assess the effects of experimental variable manipulation, and the most appropriate methods for sample acquisition and generation of the test data. The overall practicality of the project as well as the time frame for data collection and evaluation are determined at this stage in the development process.

Practical issues that may need to be addressed include the lifespan of the animal model (for chronic studies), the anticipated progression of disease in that model (to determine appropriate time points for evaluation), the amount of personnel time available for the project, and the costs associated with performing the experiments ( De Boer et al. 1975 ). If the animals are to receive chemical or biological treatments, an appropriate method for administration must be identified (e.g., per os via the diet or in drinking water [soluble substances only], by osmotic pump, or by injection). Known or potential hazards must also be identified, and appropriate precautions to minimize risk from these hazards must be incorporated into the plan. All experimental procedures should be detailed through standard operating procedures, a requirement of good laboratory practice standards ( EPA 1989 ; FDA 1987 ).

Finally, the methods to be used for data analysis should be determined. If statistical analysis is required to document a difference between experimental groups, the appropriate statistical tests should be identified during the design stage. A conclusion will be drawn subsequently from the analysis of the data with the initial question answered and/or the hypotheses accepted or rejected. This process will ultimately lead to new questions and hypotheses being formulated, or ideas as to how to improve the experimental design.

Experimental Unit

The entity under study is the experimental unit, which could be an individual animal or a group. For example, an individual rat is considered the experimental unit when a drug therapy or surgical procedure is being tested, but an entire litter of rats is the experimental unit when an environmental teratogen is being tested. For purposes of estimating error of variance, or standard error for statistical analysis, it is necessary to consider the experimental unit ( Weber and Skillings 2000 ). Many excellent sources provide discussions of the types of experimental units and their appropriateness ( Dean and Voss 1999 ; Festing and Altman 2002 ; Keppel 1991 ; Wu and Hamada 2000 ).

N Factor: Experimental Group Size

The assignment of an appropriate number of animals to each group is critical. Although formulas to determine the proper number of animals can be found in standard statistical texts, we recommend consulting a statistician to ensure appropriate experimental design for the generation of statistically significant results ( Zolman 1993 ). Indeed, the number of animals assigned to each experimental group is often determined by the particular statistical test on the basis of the anticipated magnitude of difference between the expected outcomes for each group. The number of animals that can be grouped in standard cages is a practical consideration for determining experimental group size. For example, standard 71 sq in (460 sq cm) polycarbonate shoebox cages can house up to four adult mice, so group sizes that are divisible by four will maximize group size and minimize per diem costs.

A plethora of variables (e.g., genetic, environmental, infectious agents) can potentially affect the outcome of studies performed with animals. It is therefore critical to use control animals to minimize the impact of these extraneous variables or to recognize the possible presence of unwanted variables. In general, each individual experiment should use control groups of animals that are contrasted directly to the experimental groups of animals. Multiple types of controls include positive, negative, sham, vehicle, and comparative.

Positive Controls

In positive control groups, changes are expected. The positive control acts as a standard against which to measure difference in severity among experimental groups. An example of a positive control is a toxin administered to an animal, which results in reproducible physiological alterations or lesions. New treatments can then be used in experimental groups to determine whether these alterations may be prevented or cured. Positive controls are also used to demonstrate that a response can be detected, thereby providing some quality control on the experimental methods.

Negative Controls

Negative controls are expected to produce no change from the normal state. In the example above, the negative control would consist of animals not treated with the toxin. The purpose of the negative control is to ensure that an unknown variable is not adversely affecting the animals in the experiment, which might result in a false-positive conclusion.

Sham Controls

A sham control is used to mimic a procedure or treatment without the actual use of the procedure or test substance. A placebo is an example of a sham control used in pharmaceutical studies ( Spector 2002 ). Another example is the surgical implantation of “X” into the abdominal cavity. The treated animals would have X implanted, whereas the sham control animals would have the same surgical procedure with the abdominal cavity opened, as with the treated animals, but without having the X implanted.

Vehicle Controls

A vehicle control is used in studies in which a substance (e.g., saline or mineral oil) is used as a vehicle for a solution of the experimental compound. In a vehicle control, the supposedly innocuous substance is used alone, administered in the same manner in which it will be used with the experimental compound. When compared with the untreated control, the vehicle control will determine whether the vehicle alone causes any effects.

Comparative Controls

A comparative control is often a positive control with a known treatment that is used for a direct comparison to a different treatment. For example, when evaluating a new chemopreventive drug regime in an animal model of cancer, one would want to compare this regime to the chemopreventive drug regime currently considered “accepted practice” to determine whether the new regime improves cancer prevention in that model.

Randomization

Randomization of the animals assigned to different experimental groups must be achieved to ensure that underlying variables do not result in skewed data for each experimental group. To achieve randomization, it is necessary to begin by defining the population. A homogeneous population consists of animals that are considered to share some characteristics (e.g., age, sex, weight, breed, strain). A heterogeneous population consists of animals that may not be the same but may have some common feature. Generally, the better the definition of the group, the less variable the experimental data, although the results may be less pertinent to large broad populations. Methods commonly used to achieve randomization include the following ( Zolman 1993 ):

Identifying each animal with a unique identification number, then drawing numbers “out of a hat” and randomly assigning them in a logical fashion to different groups. For example, the first drawn number is assigned to group 1, the second to group 2, the third to group 1, the fourth to group 2, and so forth. Dice or cards may also be used to randomly assign animals to experimental groups.

Using random number tables or computer-generated numbers/sampling to achieve randomization.

Experimental Protocol Approval

Animal experimentation requires IACUC approval of an animal care and use protocol if the species used are covered under the Animal Welfare Act (regardless of funding source), the research is supported by the National Institutes of Health and involves the use of vertebrate species, or the animal care program is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International ( Silverman et al. 2000 ). In practice, virtually all animal experiments require IACUC approval, which entails full and accurate completion of appropriate protocol forms for submission to the IACUC, followed by clarification or necessary modification of any procedures the IACUC requires. Approval must be obtained before the animal purchase or experimentation and is required before submission of a grant proposal by some funding agencies. If the research involves hazardous materials, then protocol approval from other intramural oversight committees or departments may also be required (e.g., a Biosafety Committee if infectious agents or recombinant DNA are to be used, or a Radiation Safety Committee if radioisotopes or irradiation are to be used).

Animal welfare regulations and Public Health Service policy mandate that individuals caring for or using research animals must be appropriately trained. Specifically, all personnel involved in a research project must be appropriately qualified and/or trained in the methods they will be performing for that project. The institution where the research is being performed is responsible for ensuring this training, although the actual training may occur elsewhere.

Pilot Studies

Pilot studies use a small number of animals to generate preliminary data and/or allow the procedures and techniques to be solidified and “perfected” before large-scale experimentation. These studies are commonly used with new procedures or when new compounds are tested. Preliminary data are essential to show evidence supporting the rationale of a proposal to a funding agency, thereby increasing the probability of funding for the proposal. All pilot projects must have IACUC approval, as for any animal experiment. As soon as the pilot study is completed, the IACUC representative will either give the indication to proceed to a full study or will indicate that the experimental manipulations and/or hypotheses need to be modified and evaluated by additional pilot studies.

Data Entry and Analysis

The researcher has the ultimate responsibility for collecting, entering, and analyzing the data correctly. When dealing with large volumes of data, it is especially easy for data entry errors to occur (e.g., group identifications switched, animal identifications transposed). Quality assurance procedures to identify data entry errors should be developed and incorporated into the experimental design before data analysis. This process can be accomplished by directly comparing raw (original) data for individual animals with the data entered into the computer or with compiled data for the group as a whole (to identify potential “outliers,” or data that deviates significantly from the rest of the members of a group). The analysis of the data varies depending on the type of project and the statistics required to evaluate it. Because this topic is beyond the scope of this article, we refer the reader to the many outstanding books and articles on statistical analysis ( Cobb 1998 ; Cox and Reid 2000 ; Dean and Voss 1999 ; Festing and Altman 2002 ; Lemons et al. 1997 ; Pickvance 2001 ; Wasserman and Kutner 1985 ; Wilson and Natale 2001 ; Wu and Hamada 2000 ).

Detection of flaws, in the developing or final experimental design is often achieved by several levels of review that are applicable to animal experimentation. For example, grant funding agencies and the IACUC provide input into the content and design of animal experiments during their review processes and may also serve as advisory consultants before submission of the grant proposal or animal care and use protocol. Scientific peers and the scientific literature also provide invaluable information applicable to experimental design, and these resources should be consulted throughout the experimental design process. Finally, scientific peer-reviewed journals provide a final critical evaluation of the soundness of the experimental design. The overall quality of the experimental data is evaluated and a determination is made as to whether it is worthy of publication. Obviously, discovering major experimental design deficiencies during manuscript peer review is not desirable. Therefore, pursuit of scientific peer review throughout the experimental design process should be exercised routinely to ensure the generation of valid, reproducible, and publishable data.

The steps listed below comprise a practical sequence for designing and conducting scientific studies. We recommend that investigators

Conduct a complete literature review and consult experts who have experience with the techniques proposed in an effort to become thoroughly familiar with the topic before beginning the experimental design process.

Ask a specific question and/or formulate an appropriate hypothesis. Then design the experiments to specifically address that problem/question.

Consult a biostatistician during the design phase of the project, not after performing the experiments.

Choose proper controls to ensure that only the variable of interest is evaluated. More than one control is frequently required.

Start with a small pilot project to generate preliminary data and work out procedures and techniques. Then proceed to larger scale experiments to generate statistical significance.

Modify original question and procedures, ask new questions, and begin again.

Barrow J . 1991 . Theories of Everything . New York : Oxford University Press .

Google Scholar

Bennett BT Brown MJ Schofield JC . 1990 . Essentials for animal research: A primer for research personnel. In: Alternative Methodologies . Beltsville : USDA National Agricultural Library University of Illinois at Chicago p 13 – 25 .

Blount RL Bunke VL Zaff JF . 2000 . Bridging the gap between explicative and treatment research: A model and practical implications . J Clin Psych Med Set 7 : 79 – 90 .

Cobb GW . 1998 . Introduction to Design and Analysis of Experiments . New York : Springer .

Cox DR Reid N . 2000 . The theory of the design of experiments. In: Monographs on Statistics and Applied Probability 86 . Boca Raton : Chapman & Hall/CRC Press .

Dean AM Voss D . 1999 . Design and Analysis of Experiments . New York : Springer .

De Boer J Archibald J Downie HG . 1975 . An Introduction to Experimental Surgery: A Guide to Experimenting with Laboratory Animals . New York : Elsevier .

Diamond WJ . 2001 . Practical Experiment Designs for Engineers and Scientists . 3rd ed. New York : Wiley .

EPA [Environmental Protection Agency] . 1989 . Good Laboratory Practice Regulations. Federal Register 40, chapter 1, part 792 .

FDA [Food and Drug Administration] . 1987 . Good Laboratory Practice Regulations. Federal Register 21, chapter 1, part 58 .

Festing MFW Altman DG . 2002 . Guidelines for the design and statistical analysis of experiments using laboratory animals . ILAR J 43 : 244 – 258 .

Holmberg P . 1996 . From dogmatic discussions to observations and planned experiments: Some examples from early aurora borealis research in Finland . Sci Educ 5 : 267 – 276 .

Keppel G . 1991 . Design and Analysis: A Researcher's Handbook . 3rd ed. Englewood Cliffs : Prentice Hall .

Kuhn T . 1962 . The Structure of Scientific Revolutions . Chicago : University of Chicago Press .

Larsson NO . 2001 . A design view on research in social sciences . Syst Prac Act Res 14 : 383 – 405 .

Lawson AE . 2002 . What does Galileo's discovery of Jupiter's moons tell us about the process of scientific discovery? Sci Educ 11 : 1 – 24 .

Lemons J Shrader-Frechette K Cranor C . 1997 . The precautionary principle: Scientific uncertainty and type I and type II errors . Found Sci 2 : 207 – 236 .

Pickvance CG . 2001 . Four varieties of comparative analysis . J Hous Built Env 16 : 7 – 28 .

Russell WMS Burch RL . 1959 . The Principles of Humane Experimental Technique . London : Methuen & Co. Ltd . [Reissued: 1992, Universities Federation for animal Welfare Herts , England .] http://altweb.jhsph.edu/publications/humane_exp/het-toc.htm .

Silverman J Suckow MA Murthy S NIH IACUC . 2000 . The IACUC Handbook . Boca Raton : CRC Press .

Spector R . 2002 . Progress in the search for ideal drugs . Pharmacology 64 : 1 – 7 .

Sproull NL . 1995 . Handbook of Research Methods: A Guide for Practitioners and Students in the Social Sciences . 2nd ed. Metuchen : Scarecrow Press .

Wasserman W Kutner MH . 1985 . Applied Linear Statistical Models: Regression, Analysis of Variance and Experimental Designs . 2nd ed. Homewood : RD Irwin .

Weber D Skillings JH . 2000 . A First Course in the Design of Experiments: A Linear Models Approach . Boca Raton : CRC Press .

Webster IW . 1985 . Starting to do research . Med J Aust 142 : 624 .

Whitcom PJ . 2000 . DOE Simplified: Practical Tools for Effective Experimentation . Portland : Productivity .

Wilson EB . 1952 . An Introduction to Scientific Research . New York : McGraw-Hill .

Wilson JB Natale SM . 2001 . “Quantitative” and “qualitative” research: An analysis . Int J Value-Based Mgt 14 : 1 – 10 .

Wu CF Hamada M . 2000 . Experiments: Planning, Analysis, and Parameter Design Optimization . New York : Wiley .

Zolman JF . 1993 . Biostatistics: Experimental Design and Statistical Inference . New York : Oxford University Press .

Abbreviation used in this article: IACUC, institutional animal care and use committee.

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1930-6180
• Print ISSN 1084-2020
• Copyright © 2024 Institute for Laboratory Animal Research
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Animal Experimentation

There are many opinions about the pros and cons of using animals in scientific research. Read the overview below to gain a balanced understanding of the issue and explore the previews of opinion articles that highlight many perspectives on animal testing.

Access Through Your library >>  

Topic Home      |      Social Issues      |      Literature      |      Lifelong Learning & DIY      |      World History

Animal experimentation topic overview.

"Animal Experimentation." Opposing Viewpoints Online Collection , Gale, 2021.

Animal experimentation, also called animal testing , has contributed to many important scientific and medical discoveries. Breakthroughs include the development of many antibiotics, insulin therapy for diabetes, modern anesthesia, vaccines for whooping cough and other diseases, the use of lithium in mental health treatments, and the discovery of hormones. Studies using animals have also led to the development of new surgical techniques and medical devices. Scientists use animals for testing the safety of chemical products, known as toxicology testing , and for evaluating the effects of radiation and biological and chemical processes. Unlike field research, which involves observing animals in their natural habitats, animal experimentation takes place at laboratories in universities, medical schools, government facilities, and commercial facilities such as those run by pharmaceutical and cosmetics manufacturers. Experiments on animals can involve testing drugs and other substances as well as performing behavioral tests such as those conducted on dogs by Russian physiologist Ivan Pavlov in the early twentieth century.

Many people object to the use of animals in scientific studies because the animals are denied their freedom and often suffer serious injury and discomfort. Other people identify certain practices used in animal studies as cruel while still recognizing the benefits of using live animals when no alternative is available. Proponents of animal experimentation maintain that these studies provide benefits to humans that cannot be achieved through other means. Conversely, critics of using animals to learn more about humans contend that the differences between humans and nonhuman species are too great for such studies to produce meaningful results. In response, proponents note that humans are not the only beneficiaries of this type of research. Many experiments are carried out to further veterinary treatments and services, improve environmental protection efforts, and better understand diseases that affect nonhuman animals and plants.

• Scientists have often used animals to learn about biology, test new surgical techniques, and observe the effects of different products on living things to determine the products' safety.
• Because humans and other animals respond differently when exposed to different substances, critics of animal experimentation have questioned the scientific value of using animals to test products intended for humans.
• Some opponents of animal experimentation contend that no scientific discovery can justify the conditions endured by animal test subjects.
• Researchers in the United States have faced difficulty obtaining lab monkeys, as several countries that previously supplied to US labs have enacted bans on such exports. A worldwide decrease in the supply of primates used in research coincided with increased demand during research for a vaccine for COVID-19 .
• The US Department of Agriculture enforces the Animal Welfare Act, which establishes regulations for the treatment of some animals used in research. While the law protects a range of species, it does not cover many of the species most commonly used in research, such as mice and fish.
• Vivisection refers to surgical experiments performed on living specimens, while dissection refers to experiments performed on dead specimens. Many companies, research institutes, and schools are exploring alternatives to such practices.
• Researchers have developed microdevices that use cell cultures to determine how different products would affect human physiology. In some cases, these devices can provide more useful and precise information than that gathered from experiments conducted on animals.

Regulations on Animal Test Subjects

Congress has enacted several pieces of legislation to regulate animal experimentation and prevent animal abuse, including the Animal Welfare Act (AWA), first passed as the Laboratory Animal Welfare Act in 1966; the Improved Standards for Laboratory Animals Act (ISLAA), passed as part of the Food Security Act of 1985; and the Health Research Extension Act of 1985, which tasks the National Institutes of Health (NIH) with establishing research standards. The AWA requires research facilities that use animals to establish an institutional committee, including at least one veterinarian and one person otherwise unaffiliated with the organization, to ensure compliance with the law. Established in 2000 as part of the NIH, the Office of Laboratory Animal Welfare (OLAW) implements federal policy and provides guidance to institutions receiving federal support.

As reported by the US Department of Agriculture (USDA), 780,070 animals were used in experiments at USDA-registered facilities in fiscal year 2018. The total includes only animals protected by the AWA and omits amphibians, birds, fish, mice, rats, and reptiles, which combined account for the majority of animals used in scientific studies. Of the animals monitored by the USDA, the most commonly used in laboratories is the guinea pig, which has been widely used for experimentation since the eighteenth century, leading it to become synonymous with a subject of any experiment. Guinea pigs, rabbits, and hamsters account for more than half of the animals in the totals reported by the USDA. The other reported test subjects include nonhuman primates, dogs, pigs, cats, and sheep.

Laboratories obtain animals for their experiments through three types of dealers: those licensed by the USDA as Class A dealers, those licensed as Class B dealers, or those not licensed at all. Class A dealers breed and raise animals for specific purposes in a closed, regulated environment. Class B dealers are less regulated and purchase or obtain animals to resell. The USDA excuses some breeders and dealers from licensing because of the type, amount, or intended use of the animals. Some states require research facilities to purchase solely from Class A dealers. Class B dealers often acquire animals from animal shelters and then sell them to research facilities.

Investigations in the 1990s revealed that some Class B dealers abducted family pets. This phenomenon led lawmakers to introduce the Pet Safety and Protection Act as an amendment to the AWA in 1996. The provision would have banned research facilities from using any dog or cat that was not obtained from a legal source. The amendment was not adopted, nor was it adopted when reintroduced in nearly every subsequent session of Congress, most recently in 2019. Despite the amendment repeatedly failing to become law, the National Institutes of Health (NIH) adopted rules in 2012 and 2014 that ended NIH funding for research involving cats and dogs from Class B or unlicensed dealers. Likewise, a provision to the Consolidated Appropriations Act of 2016 prevents the USDA from using any funds appropriated by the act to provide or renew licenses for Class B dealers. The law has effectively made it impossible for Class B dealers to obtain licenses to sell cats or dogs for research purposes. Some critics have questioned the need for the Pet Safety and Protection Act, noting that cats and dogs make up only a small portion of the animals used in experiments.

In 2017 Representative Martha McSally (R-AZ) introduced the Humane Cosmetics Act, which aims to phase out the use of animal testing in the cosmetics industry. In 2018 Representatives Mike Bishop (R-MI) and Jimmy Panetta (D-CA) followed by Senator Jeff Merkley (D-OR) introduced the Kittens in Traumatic Testing Ends Now (KITTEN) Act, which would ban the use of cats in any painful or stressful experiment, to their respective chambers of Congress. No action was taken on either the Humane Cosmetics Act or the KITTEN Act in 2019, so lawmakers reintroduced similar pieces of legislation in the subsequent session of Congress. Subsequently, in April 2019, USDA announced it would stop using cats in research. Also in 2019, lawmakers introduced the Humane and Existing Alternatives in Research and Testing Sciences (HEARTS) Act, which would prioritize federal funding for research that substituted animal subjects with alternatives. As of 2021, however, none of these bills had received a vote.

Scientists in certain fields have favored using nonhuman primates in experiments because they closely resemble humans in physiology. The National Aeronautics and Space Administration (NASA), for example, sent several nonhuman primates into space before sending astronaut Alan Shepard in 1961. Many laboratories worked with chimpanzees throughout the twentieth century. Animal behaviorists, noting the chimpanzee's intelligence and capacity for emotion, raised concerns that the use of chimpanzees in experiments amounted to torture. The Institute of Medicine deemed the use of chimpanzees in scientific research unnecessary in a 2011 report commissioned by the NIH. This report was followed by a proposal by the United States Fish and Wildlife Service (USFWS) to include captive chimpanzees, such as those used in research facilities, on the list of animals protected by the Endangered Species Act; chimpanzees in the wild had already been protected by the act since 1990. In 2013 the NIH announced its intentions to stop providing funding or granting research requests for experiments involving chimpanzees, and the USFWS proposal was finalized in 2015. Many NIH chimpanzees have since been moved to federal sanctuaries. However, scientists have chosen not to resettle many older research chimpanzees in sanctuaries because of concerns the move would worsen their health.

Though researchers have largely stopped using chimpanzees, other nonhuman primates continue to serve as research subjects. However, the policies of other countries have limited their availability. Since 2013, for example, India has banned foreign monkey exports, forcing several organizations to find new suppliers or limit their experiments. Obtaining research monkeys became increasingly difficult in 2020 when China, which had supplied more than 60 percent of research monkeys imported into the United States, instituted a ban on wildlife sales. The ban came in response to concerns that wildlife sales had contributed to the novel coronavirus disease (COVID-19) outbreak that originated in Wuhan, China, and was declared a worldwide pandemic in March 2020 by the World Health Organization (WHO). Though many animal rights activists and public health officials applauded China's decision, the ban had the unintended effect of reducing the supply of lab monkeys at a time when demand significantly increased as medical researchers and pharmaceutical companies sought to develop a vaccine for COVID-19. Like vaccines for other diseases, the COVID-19 vaccines available in the United States as of March 2021 were approved for emergency use based in part on results of experiments on animals including mice, rats, hamsters, and monkeys.

Alternatives to Animal Testing

Animal rights advocates and members of the scientific community have pushed for the use of alternatives to animal experimentation. The pursuit of alternatives has largely centered on concepts first introduced in 1959 by British zoologists W. M. S. Russell and R. L. Burch in The Principles of Humane Experimental Technique . Russell and Burch framed their proposal around three Rs. They suggested that experimentation should replace animal subjects with something else, such as nonsentient material or less sentient animals; reduce the number of animal subjects used experimentally while increasing the amount of data obtained; and refine living conditions and experimental procedures for animal subjects to reduce pain and discomfort.

The NIH established the Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) in 2000 as part of the National Toxicology Program Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM) to promote and regulate alternatives to animal testing. In addition to government programs, animal rights advocacy groups such as People for the Ethical Treatment of Animals (PETA) and the American Fund for Alternatives to Animal Research (AFAAR) also contribute funding to develop alternative research methods. AAALAC International, formerly known as the Association for Assessment and Accreditation of Laboratory Animal Care, distributes up to four $5,000 prizes each year to researchers that make significant contributions to improving the nature of animal research. The awards are a component of the organization's Global 3Rs Awards program, named for the principles put forth by Russell and Burch.

Companies and research facilities, however, can be slow to adopt alternatives. In 2009, for example, the Organisation for Economic Cooperation and Development (OECD) approved two alternatives to the Draize test, a research method that involves applying chemicals directly to the eyes or skin of animals, typically rabbits. The test is widely condemned by animal rights activists. In 2020 university researchers in the United Kingdom announced a method that they determined to be both cheaper and more ethical than the Draize test, as flatworms served as a substitute for rabbits. As of 2021, despite the availability of these alternatives, scientists continue to perform the Draize test, arguing that no single test has proven able to replicate the full benefits of the Draize test.

Technological advances have enabled scientists to perform many experiments without using live animals. Invasive animal experimentation that involves performing surgery on a living animal can be referred to as vivisection , as opposed to dissection , which is surgery performed on a deceased animal. The term vivisection, however, is typically used by opponents of animal experimentation and avoided by scientists. Researchers have developed ways to obtain data without using live specimens by experimenting on cells and tissues rather than the entire living organism; these procedures are referred to as in vitro experiments. In many in vitro experiments, human cells and tissues can be used. Proponents argue that this method produces data that is more relevant to human safety. Critics of in vitro methods argue that operating on a live animal provides more accurate data because the effects on the entire organism can be observed.

US schools began incorporating dissection into biology instruction in the 1920s, with the practice becoming widespread by the 1960s. A 2014 survey conducted by the National Association of Biology Teachers (NABT) found that 84 percent of biology teachers and 76 percent of biology students were using dissection in the classroom. Many of the responding teachers, however, reported that their schools were shifting away from dissection and pursuing alternatives such as virtual dissection programs, 3D models, and videos, largely in response to student requests. Educators also reported using these alternatives alongside traditional hands-on dissection. In 2019 the NABT reaffirmed its belief that students should have access to living and formerly living specimens and that nonanimal alternatives may not provide students with the most comprehensive understanding of life science. However, the NABT stresses the importance of teachers educating students about maintaining professional and ethical standards in animal research.

In the early 2000s, researchers began developing microdevices referred to as organs-on-a-chip (OOCs), which use cell cultures to imitate a human organ and determine how that organ would respond to different chemicals and other stimuli. OOCs are approximately the size of a deck of playing cards and have been developed to imitate lungs, hearts, kidneys, skin, eyes, and entire organ systems. In 2018 researchers successfully tested OOCs that imitated interconnected organ systems and could produce data for twenty-eight days, indicating that a microdevice could likely support an entire "human on a chip." In some cases, computer models can simulate the effects of diseases and medicines on the human body with greater accuracy than animal subjects. Research methods that substitute computer models for live animals are referred to as in silico experiments.

Critical Thinking Questions

• In your opinion, why have federal lawmakers delayed holding votes on legislation introduced in the late 2010s that would have expanded protections for animals used in research?
• Would you support a ban on the use of dissection in high school biology classrooms? Why or why not?
• Under what conditions, if any, do you think scientists should be allowed to use animal subjects in their research? Explain your answer.

Extremist Activism

Despite efforts to reduce the number of animals used in scientific studies and minimize the pain and distress that animal subjects experience, some animal rights activists believe that the benefits of animal experimentation do not justify the cruelties involved. Some extremist groups of activists calling for an end to all animal testing have engaged in criminal activity to prevent animals from being used in experiments. In the late 1970s, radical animal rights groups began targeting companies and research facilities, using terrorist strategies to disrupt these industries and promote their extremist platform. These activists were sometimes called "ecoterrorists" by federal authorities and included members of radical groups such as the Animal Liberation Front (ALF) and Stop Huntingdon Animal Cruelty (SHAC).

To protect research companies and other commercial enterprises vulnerable to animal rights violence, Congress passed the Animal Enterprise Protection Act of 1992 and the Animal Enterprise Terrorism Act of 2006. Critics of these laws note that both bills received support from biomedical and agribusiness lobbying groups. Additionally, critics note that both laws include language that criminalizes activities protected by the First Amendment, such as picketing and leading boycotts, if they interfere with a company's ability to make money. In 2015 two animal rights activists, Kevin Johnson and Tyler Lang, challenged the constitutionality of the Animal Enterprise Terrorism Act after they were charged with violating the act for vandalizing a mink farm and setting hundreds of animals free in 2013. However, the United States Court of Appeals for the Seventh Circuit ruled that the law was constitutional in November 2017. By April 2021, this type of extreme action to stop animal experimentation has become rare, with no major events reported in the United States since 2013.

More Articles

Government regulations will encourage alternatives to animal experimentation.

“There was a time when dosing and contaminating animals with often toxic levels of chemicals was horrible for them but imperative for human health and safety.”

The Times Editorial Board determines the perspectives and positions of the news organization.

In the following viewpoint, the authors contend that the recent overhaul of the Toxic Substances Control Act will lead to a reduction in the number of animals subjected to experimentation. Revisions to the law, the authors maintain, will encourage companies to employ alternative methods for gathering data and work with other companies to reduce instances of duplicated experiments. The authors argue that advances in technology and a new willingness among companies to cooperate with one another have eliminated the need to test products on animals to ensure they are safe for humans to use.

Using Monkeys for Research Is Justified—It’s Enabling Treatments that Would Be Otherwise Impossible

“I am confident that the next 50 years will see wonderful progress in treatments for these terrible disorders and primate research will be central to this effort.”

Stuart Baker is a professor of movement neuroscience at Newcastle University in the United Kingdom.

In the following viewpoint, Baker argues that the expanded use of primates and other animals in experiments is necessary to find a cure to challenging diseases like neurological disorders among the elderly. Baker refutes the argument of critics that animals used in research are subjected to extreme suffering and contends that researchers follow state-of-the-art surgical procedures commonly used on humans. As a researcher himself, Baker maintains that the primates he used in his experiments willingly cooperated and did not exhibit any signs of stress. For the author, the use of animals in scientific pursuits is essential for alleviating suffering among human beings.

The Grim Good of Animal Research

"Experimenting with animals before testing on people is a crucial human rights protection required by the famous Nuremberg Code."

In the following viewpoint, Wesley J. Smith argues that research on animals has been indispensable in developing ways to treat human disease. No one likes the idea of experimenting on animals, he says, and efforts are being made to reduce it to a minimum; however, there is no other way to do the necessary research and check the safety of new drugs. Medical treatments have to be tested on living organisms; if not on animals, then on humans, which in Smith's opinion would be an atrocity. Smith is a senior fellow for the Discovery Institute's program on human exceptionalism. He also consults with the Patients Rights Council and the Center for Bioethics and Culture.

Results from Research on Animals Are Not Valid When Applied to Humans

"Animal advocates, as well as many scientists, are increasingly questioning the scientific validity and reliability of animal experimentation."

In the following viewpoint, the American Anti-Vivisection Society (AAVS) declares that experimentation on animals is not a valid means of testing treatments for human disease. The AAVS maintains that animal studies do not reliably predict human outcomes, that most drugs that appear promising in animal studies go on to fail in human clinical trials, and that reliance on animal experimentation can delay discovery. In the opinion of the AAVS, animals are used in medical research more from tradition than from evidence of scientific value. The AAVS is a nonprofit animal advocacy organization dedicated to ending experimentation on animals in research, testing, and education.

Looking for information on other topics?

Access Through Your Library >>

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Ethical issues in research: Human and animal experimentation

Affiliations.

• 1 Department of Pediatric Surgery, CHU Pellegrin-Enfants, Bordeaux, France. Electronic address: [email protected].
• 2 Division of Urology/Department of Research, Connecticut Children's Medical Center, Hartford, USA.
• 3 Service de Chirurgie et Urologique Pédiatrique Hôpital Lapeyronie, CHU de Montpellier, Université de Montpellier, France.
• PMID: 29477694
• DOI: 10.1016/j.jpurol.2017.12.012

Keywords: Animal; Ethics; Human; Research.

PubMed Disclaimer

Similar articles

• Biomedical research ethics. Iltis AS. Iltis AS. J Med Philos. 2002 Oct;27(5):515-22. doi: 10.1076/jmep.27.5.515.10319. J Med Philos. 2002. PMID: 12461708 No abstract available.
• Ethical issues in research: Human and animal experimentation. Harper L, Herbst KW, Kalfa N. Harper L, et al. J Pediatr Urol. 2018 Jun;14(3):286. doi: 10.1016/j.jpurol.2017.10.019. Epub 2017 Dec 2. J Pediatr Urol. 2018. PMID: 29428361 Review. No abstract available.
• Handbook of medical ethics: ethics in biomedical research. [No authors listed] [No authors listed] Issues Med Ethics. 1996 Jul-Sep;4(3):90-3. Issues Med Ethics. 1996. PMID: 15011683 No abstract available.
• Teaching ethics to basic scientists: suggestions for greater curricular clarity. Root Wolpe P. Root Wolpe P. Am J Bioeth. 2002 Fall;2(4):62-3. doi: 10.1162/152651602320957646. Am J Bioeth. 2002. PMID: 12762933 No abstract available.
• Animal models in biological and biomedical research - experimental and ethical concerns. Andersen ML, Winter LMF. Andersen ML, et al. An Acad Bras Cienc. 2019;91(suppl 1):e20170238. doi: 10.1590/0001-3765201720170238. Epub 2017 Sep 4. An Acad Bras Cienc. 2019. PMID: 28876358 Review.
• [From the laboratory to the clinic in acute ischaemic stroke. In vitro and in vivo experimental models]. Pradillo JM, García-Culebras A, Cuartero MI, Peña-Martínez C, Moro MA, Lizasoain I, Moraga A. Pradillo JM, et al. Rev Neurol. 2022 Nov 1;75(9):283-293. doi: 10.33588/rn.7509.2022268. Rev Neurol. 2022. PMID: 36285448 Free PMC article. Review. Spanish.
• Current Methods in the Study of Nanomaterials for Bone Regeneration. Tanaka M, Izumiya M, Haniu H, Ueda K, Ma C, Ueshiba K, Ideta H, Sobajima A, Uchiyama S, Takahashi J, Saito N. Tanaka M, et al. Nanomaterials (Basel). 2022 Apr 2;12(7):1195. doi: 10.3390/nano12071195. Nanomaterials (Basel). 2022. PMID: 35407313 Free PMC article. Review.
• Electro-mechanical Lung Simulator Using Polymer and Organic Human Lung Equivalents for Realistic Breathing Simulation. Pasteka R, Forjan M, Sauermann S, Drauschke A. Pasteka R, et al. Sci Rep. 2019 Dec 24;9(1):19778. doi: 10.1038/s41598-019-56176-6. Sci Rep. 2019. PMID: 31874980 Free PMC article.

Publication types

• Search in MeSH

LinkOut - more resources

Full text sources.

• Elsevier Science

Other Literature Sources

• scite Smart Citations
• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Last updated 10th July 2024: Online ordering is currently unavailable due to technical issues. We apologise for any delays responding to customers while we resolve this. For further updates please visit our website https://www.cambridge.org/news-and-insights/technical-incident

We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings .

Login Alert

• > Journals
• > Cambridge Quarterly of Healthcare Ethics
• > Volume 31 Issue 1
• > Animal Research that Respects Animal Rights: Extending...

Article contents

Introduction, the rights of animals, requirements for ethical research with humans, extending research requirements for humans to animals, animal research that respects animal rights: extending requirements for research with humans to animals.

Published online by Cambridge University Press:  20 January 2022

The purpose of this article is to show that animal rights are not necessarily at odds with the use of animals for research. If animals hold basic moral rights similar to those of humans, then we should consequently extend the ethical requirements guiding research with humans to research with animals. The article spells out how this can be done in practice by applying the seven requirements for ethical research with humans proposed by Ezekiel Emanuel, David Wendler, and Christine Grady to animal research. These requirements are (1) social value, (2) scientific validity, (3) independent review, (4) fair subject selection, (5) favorable risk–benefit ratio, (6) informed consent, and (7) respect for research subjects. In practice, this means that we must reform the practice of animal research to make it more similar to research with humans, rather than completely abolish the former. Indeed, if we ban animal research altogether, then we would also deprive animals of its potential benefits—which would be ethically problematic.

The ethics of animal experimentation is a controversial topic, currently receiving much attention from both animal researchers and bioethicists. Some argue that animal research is permissible if currently implemented research requirements are met and the research protocol is approved by an independent animal ethics committee (AEC; sometimes also called Institutional Animal Care and Use Committee—IACUC). Others, especially defenders of animal rights such as Tom Regan and Alasdair Cochrane, defend the view that harmful animal research is immoral and should be abolished. However, both philosophers admit that nonharmful animal experimentation may be ethically justified. For example, Regan states that “[t]he rights view is not against research on animals, if this research does not harm these animals or put them at risk of harm.” Footnote 1 And Cochrane claims that “[a]s such, animals have no moral right not to be used for certain purposes if their well-being is respected, and that includes their use in experimentation. If scientists experiment on animals in ways that cause no pain and that do not end in death, then such experiments are permissible.” Footnote 2 However, neither Regan nor Cochrane flesh out what animal research that respects animal rights would be in practice.

This article aims to close this gap in the animal rights literature by arguing that if animals have the same moral worth as humans, along with inviolable rights, then ethical research requirements and principles regulating research with humans should be extended to them as well. Indeed, research with humans—even with groups who cannot speak for themselves—is permitted, so long as some basic requirements are met. If groups unable to consent to their participation in studies were entirely excluded from research, they could not gain from its potentially beneficial discoveries. The same holds for nonhuman animals: research with them should not be condemned outright, but rather conducted in a different way—viz., so that they can also potentially benefit from research results.

To be sure, there is controversy about the standards required for ethical research with humans. In order to make my argument, I rely on the list of requirements that all research with humans must meet presented by Ezekiel Emanuel, David Wendler, and Christine Grady. In their article “What Makes Clinical Research Ethical?,” they list seven requirements, including (1) social value, (2) scientific validity, (3) independent review, (4) fair subject selection, (5) favorable risk–benefit ratio, (6) informed consent, and (7) respect for research subjects. Footnote 3 I outline how these requirements can be applied to the case of animal research in practice; and argue that if animals have basic rights, then much higher research standards than the ones currently implemented have to be met in order for animal research to become ethically legitimate.

The thesis defended here—standards implemented for research with humans should be extended and applied to animals—is not entirely new (see, e.g., Hope Ferdoswian and Chong Choe). Footnote 4 Over the last 15 years, many authors have advocated extending the principles and requirements used in research with humans to animals. However, most authors have respectively focused on a single research requirement, such as respect for autonomy, Footnote 5 , Footnote 6 harm–benefit ratio, Footnote 7 , Footnote 8 respect, Footnote 9 or vulnerability. Footnote 10 , Footnote 11 The aim is to present a more complete picture here.

Two recent publications present more comprehensive accounts of the research principles for animal research. In their article, “A Belmont Report for Animals,” Hope Ferwodsian, Syd L. Johnson, Jane Johnson, Andrew Fenton, Adam Shriver, and John Gluck argue that the key ethical principles of the Belmont Report—that is, respect for persons and their autonomy, beneficence, justice, and special protection for vulnerable individuals and populations—should be binding for animal research. Footnote 12 The authors call for an internationally binding document for animal research invoking these principles. Their project is important and laudable, and I agree with most of their conclusions. However, by focusing on the research requirements proposed in the Belmont Report, they leave aside some elements that are constitutive of ethical animal research. Their account should thus be complemented and refined with further principles.

Tom Beauchamp and David DeGrazia recently presented a more detailed account of ethical principles for animal research. Footnote 13 In their book, Principles of Animal Research Ethics , they outline six moral principles that should govern animal research—three related to social benefit and three related to animal welfare:

1) the principle of no alternative method;

2) the principle of expected net benefit;

3) the principle of sufficient value to justify harm;

4) the principle of no unnecessary harm;

5) the principle of basic needs; and

6) the principle of an upper limit to harm.

Beauchamp and DeGrazia claim that these principles “can be accepted by all parties who are enthusiastic about the history and promise of animal research and all parties who are enthusiastic about vigorous protection of animal research subjects’ welfare—without sacrifice of anyone’s basic commitments.” Footnote 14 That is, they seek a compromise between proponents of animal research and those wanting to protect animal welfare, limiting their framework to those values that most people are likely to endorse. Footnote 15 Thus, their strategy is to propose principles on which animal research proponents and defenders of the animal cause can converge, with a view to forging a pragmatic compromise that will substantially improve animal research. The upshot is that they provide a rather pragmatic framework for ethical animal research, but are not concerned with the question of what ethical research would look like if animals had fundamental rights.

As a result, there is still no full and detailed account that fleshes out what animal research should look like if animals have fundamental rights. Accordingly, the aim here is to outline the principles that should govern animal research if we accept the premise that animals have basic rights. Footnote 16

The argument proceeds as follows: First, I discuss the fundamental rights of animals. Since the claim of this paper is a conditional one—viz., if animals have moral rights, then the ethical requirements, standards, and principles governing research with humans should be applied to animal research—a clear understanding of animals’ rights is needed. If animals have, for example, a right to live or a right to bodily integrity, this will influence what researchers are allowed to do with them. That is, the rights of animals are listed that should be taken into consideration when conducting research. Second, the standards guiding research with humans are examined, thereby relying on the requirements outlined by Emanuel, Wendler, and Grady. Lastly, it is shown how their seven requirements can be applied to animal research in practice, and it is outlined what this would mean for the future of animal experimentation.

In the last 40 years, many animal ethicists have argued that not only do animals matter morally, but they also have the same inherent worth or moral status as humans. This proposition was often advanced by rejecting speciesism. The term “speciesism” was originally introduced by Richard Ryder and taken up by Peter Singer, who presented the first detailed analysis in Animal Liberation : “Speciesism—the word is not an attractive one, but I can think of no better term—is a prejudice or attitude of bias in favor of the interests of members of one’s own species and against those of members of other species.” Footnote 17 That is, speciesist behavior entails “an unjustified disadvantageous consideration or treatment of those who are not classified as belonging to one or more particular species.” Footnote 18 Correspondingly, antispeciesists claim that a discrimination based merely on species membership is morally problematic, as species membership is morally irrelevant. Thus, if one aims for a non-speciesist treatment of animals, one has to consider these beings, their interests, and their rights in an unbiased way.

In what follows, I do not argue in favor of animal rights per se. For the sake of the argument, I start from the premise that animals have inviolable moral rights, similar to humans’. The aims in what follows are threefold: first, to show that if animals have rights, then a complete halt to animal research is unnecessary; second, to specify the fundamental rights of animals that are relevant to animal research; and third, to adumbrate what animal research could look like if we were to take animals’ rights seriously.

A useful starting-point to determine animals’ rights are the “Five Freedoms of Animal Welfare” developed by the United Kingdom Farm Animal Welfare Council (FAWC) in 1979. They include:

1) Freedom from hunger and thirst—by ready access to fresh water and a diet to maintain full health and vigor.

2) Freedom from discomfort—by providing an appropriate environment including shelter and a comfortable resting area.

3) Freedom from pain, injury, or disease—by prevention or rapid diagnosis and treatment.

4) Freedom to express normal behavior—by providing sufficient space, proper facilities and company of the animal’s own kind.

5) Freedom from fear and distress—by ensuring conditions and treatment which avoid mental suffering. Footnote 19

These Five Freedoms can be deemed individually necessary and jointly sufficient as a framework for the analysis of animal welfare. If all these requirements are fulfilled, this should lead to a high overall level of animal welfare. Footnote 20

Furthermore, the Five Freedoms can be rephrased as rights. This presupposes that the Five Freedoms involve corresponding duties, which does seem to be the case: farmers are required to respect the Five Freedoms and thus their animals’ basic needs. Footnote 21 Likewise, Clare McCausland has argued that the Five Freedoms can be regarded as welfare rights. Footnote 22 Steven McCulloch concludes: “Taken together, FAWC’s mixed ethical approach, together with the correlative nature of rights and duties, suggests that FAWC’s prescription is very close to a recommendation of moral rights for farm animals.” Footnote 23 Accordingly, animals can hold rights commensurate with their welfare interests covered by the Five Freedoms.

Although the Five Freedoms were originally developed for farmers and animal husbandry, they can also be applied to animal research—implying that researchers have a duty to respect the Five Freedoms. Research animals depend exclusively on laboratory staff members for meeting their basic needs, which results in claims for the provision of food and entertainment possibilities. That is, research animals have the right to live free from hunger and thirst, discomfort, pain, distress, and so forth, and they have the right to pursue species-typical behavior. In practice, this means that laboratory stuff and researchers must provide animals with enough space, companionship (where indicated) and possibilities for entertainment, to ensure their physical and mental well-being.

One aspect neglected by the Five Freedoms is animals’ deaths. Most currently implemented animal research guidelines are concerned with minimizing animal suffering, and animals’ deaths before, during, and after experiments is less frequently discussed as an ethical issue. However, according to many animal ethicists, not only animals’ welfare, but also their deaths matter morally. Death forecloses future opportunities for the satisfaction of interests and is thus a harm (by deprivation) to sentient animals. That is, although most sentient animals do not have a concept of death (similarly to infants or severely cognitively disabled humans), their life has a value for them: they have an interest in continued existence, insofar as it allows them to live through future pleasant experiences. Footnote 24 , Footnote 25 , Footnote 26 , Footnote 27 Moral agents have thus a pro tanto duty not to end animals’ lives unnecessarily and prematurely. Therefore, we can add the freedom to continue living one’s life to the list of the Five Freedoms.

If we phrase these freedoms in terms of rights, this leads us to the following list:

1) Right to be free from hunger and thirst;

2) Right to be free from discomfort;

3) Right to be free from pain, injury, or disease;

4) Right to be free to express normal behavior;

5) Right to be free from fear and distress;

6) Right to continued existence.

If one accepts that animals have these rights, there are consequences for their permissible use in medical research. However, as will be argued in what follows, these rights do not imply that research with animals is always ethically problematic. Many human research subjects find themselves in situations similar to those of research animals. For example, they live in extreme dependency on others (as in the case of prisoners or small children), or they are unable to understand what medical research is so as to consent to it (like severely mentally handicapped individuals or infants). Nevertheless, research with them is not categorically forbidden. Rather, if special protective measures are taken and certain rules are respected, research with them may be ethically permissible. This allows them, in turn, to profit from the potential scientific benefits of the studies in which they are involved. The same principle should apply to research with animals.

The objective of medical research is to promote society’s interest in health, knowledge, and well-being at large. In order to avoid the exploitation of research participants in the pursuit of these aims, protective ethical requirements for research have been put in place. These research requirements should be sensitive to humans’ basic rights , such as our right to bodily integrity and our right to respectful treatment. Furthermore, medical research should not override individuals’ basic rights for the sake of the majority.

In addition, some requirements in place for studies on humans are concerned with the form of the research, rather than the basic rights of the enrolled participants. Scientific validity is an example: research that does not meet this requirement is void, as the results are useless from a scientific perspective. The same applies to independent review by Institutional Review Boards (IRBs). IRBs ensure that the study pursues an important aim, that the methodology is sound, and that research participants are chosen fairly (not due to their membership of a socially salient group, for example). Thus, research should be governed by the principles of justice and impartiality.

After examining different research guidelines and medical research codes, Emanuel, Wendler, and Grady list seven requirements that should guide all research with humans:

1) social value;

2) scientific validity;

3) independent review;

4) fair subject selection;

5) favorable risk–benefit ratio;

6) informed consent; and

7) respect for research subjects. Footnote 28

In other words, research should be socially or scientifically useful , address important societal and scientific questions, and yield valuable results for society as a whole. The results should be methodologically, statistically, and scientifically sound; that is, accepted scientific principles and methods should be used in order to produce reliable and valid results. Furthermore, the results should be described in a comprehensible way that allows for the study to be replicated. Independent review requires that the study protocol be reviewed and approved by an independent committee charged with investigating whether the study addresses a scientifically important question and is thus indispensable. It also determines whether the study’s design is suitable for achieving the study’s aims and whether the risk–benefit ratio is justified. Fair subject selection demands that justice prevail when choosing the group with which research is conducted. For example, vulnerable or stigmatized groups should not be targeted for high-risk research if the research could also be conducted with other groups. That is, researchers should choose the research participants based on the study’s goals—not simply because of the greater convenience or availability of certain individuals who are willing to take high risks (e.g., due to terminal disease or poverty). Favorable risk–benefit ratio requires a just distribution of burdens and benefits: risks should be minimized and be in proportion to the expected benefits of the study (for the individual concerned and for society at large). Informed consent requires that research subjects be aware that they are participating in a study, that they understand the concomitant risks, and that they consent to their participation only after having been given all relevant information. If research subjects cannot consent due to the nature of the study (e.g., patients in emergency rooms or in a coma) or because they lack the relevant cognitive capacities for giving informed consent, then consent by proxy in the participants’ best interest or in accordance with the presumed values of the individual is required. Furthermore, the risk should be minimized to an acceptable level in such cases. Finally, respect for research subjects means that participants are permitted to withdraw from the study at any time, that their privacy is protected by confidentiality, that they are informed of newly discovered risks, benefits and the study’s results, and that their welfare is maintained throughout the study. In the following, I argue that these same requirements should be extended to research animals, and I show how this can be done in practice.

If we want to respect the rights of animals outlined above—e.g., to be free from discomfort, disease, fear, distress and pain, to express normal behavior, and to enjoy continued existence—this precondition restricts what we are consequently allowed to do with them in research. Inducing diseases in animals, for example, would no longer be permissible. Experiments that cause suffering or lead to the death of the animals would be ethically problematic. And so would experiments that severely restrict animals’ opportunity to pursue species-typical behavior, such as maternal deprivation studies.

However, these restrictions do not rule out other forms of nonharmful animal research. Animals cannot speak up for themselves and give informed consent to their participation in medical research, as they lack the requisite cognitive capacities. For this reason, they share similarities with many groups who cannot consent to research, such as infants, persons in a coma or severely cognitively disabled individuals. On the other hand, these groups should not necessarily be excluded from research, since this would deprive them of all the potential benefits that the research could bring to them. Footnote 29 Indeed, we already permit research that involves little or no risk and that is not harmful with groups unable to consent, so long as there is consent by proxy in their best interest. If animals have basic rights similar to those of humans, then the same principle should apply to them: nonharmful and low-risk research with them should be permissible, since a complete abolition of animal research would deprive animals of research results that are beneficial to them individually, or to their species.

Emanuel, Wendler, and Grady lay out ethical research requirements for studies with diverse populations, including vulnerable groups. If animals have rights similar to humans’, then we have good reasons to extend these requirements to research animals. Here, I outline how this can be done in practice. Note that three conditions are already required for animal research in most countries, namely social and scientific value, scientific validity, and independent review. Footnote 30 While necessary, these requirements are insufficient for ethical animal research. Further elements are needed that (i) guarantee animals’ welfare throughout studies and (ii) provide formal requirements for any research with sentient beings, regardless of species.

Social Value

Let us turn first to social and scientific value. As in the case of research with humans, research with animals should only be conducted if it addresses important societal and scientific questions, yields valuable results for society at large, generates new knowledge, or replicates previous results. Most animal research studies start from the assumption that the expected value should only or primarily benefit humans. However, if animals have fundamental rights, this is problematic: if studies are pursued on and with animals, then the benefits should also be useful to individual animals or their species, not solely to humans. If animal research is solely responsive to humans’ health priorities, then animals carry the undue burden of being used in research from which they do not benefit, which is speciesist. Yet many domesticated animals form part of our society, as humans live with them and benefit from inter-species relationships. Correspondingly, these animals’ interests and health priorities should also be considered in the common good. In practice, this means that when establishing research protocols, the beneficiaries of the study’s results must be identified. If they are solely a small and already over-represented group, or if the research does not address any health priority of a large group, then the study’s aims and the allocation of resources to this end should be reconsidered.

A problem concerning social and scientific values arises due to the lack of obligatory registries for past and ongoing animal experiments, as is required for research with humans. Often, negative results from animal research are not published. This may lead some research groups to undertake studies that were already conducted by other colleagues, but not published. That is, some of the studies conducted on animals fail to generate new knowledge. Although some administrative and practical difficulties would have to be surmounted to establish such registries, Footnote 31 they may be useful for preventing the multiplication of already-conducted experiments which did not result in any new knowledge. That is, such publicly accessible registries could reduce publication bias. Footnote 32 , Footnote 33 , Footnote 34 , Footnote 35

Scientific Validity

Scientific validity is commonly accepted as an important requirement for animal research—not only for scientific reasons, but also for ethical ones:

[…] if poorly conducted studies produce unreliable findings, any suffering endured by animals loses its moral justification because their use cannot possibly contribute towards clinical benefit. Non-publication of animal studies is similarly unethical because the animals involved cannot contribute towards the accumulation of knowledge and because non-publication may result in further, unnecessary animal and human experiments. Footnote 36

That is, a sufficient number of animals should be enrolled in the study to yield reliable and statistically sound results; in the published articles, experiments need to be described in a comprehensible way, enabling other researchers to replicate the study. This presupposes a detailed description of the experiments—including the sex, age, and health-status of the animals involved, along with the method and details of the statistical analysis. To avoid confirmation bias, the studies should be blinded and randomized.

The requirement of scientific validity is not always perfectly respected—in research with humans and with animals—which is why some researchers talk about a “crisis of reproducibility” in many fields, including animal research. Footnote 37 , Footnote 38 , Footnote 39 If research with animals is conducted, it should be ensured by the researchers, AECs, and journal editors that the methods and statistical analysis are well described and sound. For example, this can be done by offering researchers better training about methods and potential biases, as well as by making it obligatory for all articles accepted for publication and involving animal research to follow the ARRIVE [Animals in Research: Reporting In Vivo Experiments] guidelines for reporting animal research. Footnote 40

Independent Review

A further requirement for research with humans is independent review by IRBs. This is also an established requirement for animal research in most countries. AECs evaluate study protocols and decide whether the research should be approved. However, there are several issues with the evaluation processes for animal research. First, AECs are not yet obligatory in all countries. Second, AECs do not necessarily consist of independent evaluators, as required for IRBs. For example, in the United States, many AECs may be biased because they consist of animal researchers and veterinarians who themselves undertake animal research and thus depend on it for their own research Footnote 41 ; in consequence, they may not be impartial in their evaluations. Third, different AECs often apply divergent standards when it comes to approving the very same studies. Footnote 42 Fourth, AECs frequently do not follow international guidelines, but national ones; hence, approval by an AEC in one country may imply very different protection standards than in another. Finally, there is evidence that the reporting of methods in research protocols is often insufficient, but nevertheless approved by AECs due to implicit confidence rather than to evidence of scientific rigor. Footnote 43

However, this does not mean that AECs are problematic per se and should be abandoned. Rather, they should be reformed. Both AECs and IRBs fulfill the important role of evaluating studies’ aims and methods as well as approving the involved research populations. In order to adequately fulfill this role, AECs need clearer criteria for ethical animal research. That is, binding international guidelines for ethically acceptable research with animals should be established—as is already the case for research with humans. Such guidelines would make the evaluation process less arbitrary. Furthermore, AEC members should be completely independent of the research institution in order not to be unduly influenced or biased, and it should be ensured that at least one trained ethicist is involved in the evaluation process. Footnote 44 Finally, if animals have fundamental rights, then AECs should consider these in a nonspeciesist way. This argument is further developed in the next section.

Fair Subject Selection

Currently, animals are exposed to more of research’s risks than humans, while not expecting any of its benefits for themselves. As Chong Choe Smith writes: “[…] nonhuman animals bear a disproportionate share of the burdens of research without a showing of sufficient justification—for example, a showing that there are morally relevant and significant differences to justify the ethical use of nonhuman animals in research.” Footnote 45 This may especially concern animal species we know little about. A lack of knowledge about some species’ needs or capacity to suffer (e.g., insects) may lead to their interests’ being summarily discarded or simply ignored. Researchers may be tempted to use these animals more often or in high-risk research which supposedly does not afflict this group, thereby placing an unduly heavy burden on them.

The same applies to research that is inconsistently regulated, or not at all. Some species (such as rats, mice, fish, birds, or amphibians) are not always covered by animal welfare regulations in certain countries (e.g., the United States). However, since animals of the same species presumably do not differ regarding their experiential welfare, international guidelines should protect the interests of all animals of the same species in similar ways all around the world. That is, binding international guidelines for ethical animal research are needed—ones that require similar standards for the same species of sentient animals, regardless of where the research is being conducted. Having such internationally recognized, clear-cut criteria for ethical animal research would, in turn, also simplify the evaluation process for AECs.

In principle, researchers should always choose the best model for their research. In practice, however, some species, such as rodents, may be more frequently sought out for studies because they are easily available and do not demand a huge investment of time and money. However, if they are an inappropriate model for the condition being investigated, or if they are not afflicted by the condition being investigated, then the study should not be conducted with them.

Hence, to avoid overburdening animals in research, it must be ensured that animals are only included in research that serves them individually or their species. That is, AECs would need to keep an eye on the animal population chosen for study purposes, and demand a justification for why a specific species was singled out. In practice, this means that AECs should treat animals in research in ways similar to other society members who cannot express themselves for or against their participation in research, such as persons in a coma, infants, or severely cognitively disabled individuals. In the latter type of cases, IRBs have to check whether enrolling this group is necessary for achieving the study aims, or whether these particular groups were chosen due to mere convenience, for example. The same should apply to research animals, as their situation—dependency and inability to speak out for themselves—is similar to these populations’.

Favorable Risk–Benefit Ratio

We refrain from conducting studies with human research subjects if the harm and risk involved is too high for them—even if the results obtained would be highly beneficial to society at large. Neither do we accept the death of human research subjects as a normal consequence of a study. That is, studies with humans must have a favorable risk–benefit ratio. If animals have fundamental rights, then the same principle should apply to research with them. So far, however, animals often do not benefit from the research undertaken on them: the results obtained usually serve the human species exclusively.

Note, though, that research with humans may sometimes be permissible even if the research subjects themselves do not benefit from it, as long as the results obtained would be useful to the general population. Yet this is only the case if the research participants can consent to the study and are informed about the concomitant risks, or if the risk threshold is low so that participation is permissible without direct consent, that is, if the participants are unable to give informed consent and participation is in their interest. The acceptable risk-threshold in the case of humans unable to give informed consent remains disputed in the literature. Footnote 46 , Footnote 47 Defining a risk-threshold for animals is not the aim here and would be a task for separate project. However, what can be noted is that there should be some upper risk threshold for animal research, as is the case of research with humans:

One way of thinking about problems of upper limits is that if it is not justified to exceed fixed upper levels of pain, suffering, and distress with nonconsenting human subjects in nontherapeutic research, and if animal subjects are relevantly similar to human subjects in the relevant respects, then exceeding the same levels of pain, suffering, and distress would likewise not be justified in the use of animals in research—or at least a justification would be required to show why what is unjustified with human subjects is justified with animal subjects. When human interests and animal interests are relevantly similar and their welfare is contingent on not being constrained, coerced, deprived of basic needs, and placed in pain or terror, it is difficult to see what, if anything, would justify treating the interests of animals as dissimilar to human interests. Footnote 48

Although to date only few guidelines, such as the Preamble 23 of Directive 2010/63/EU prescribe an upper limit and a favorable risk–benefit ratio for animals, Footnote 49 it should become a requirement in all animal research guidelines.

Informed Consent

The extension of informed consent to research animals is quite frequently discussed in the literature. Hope Ferdowsian and Chong Choe, for example, state: “[…] although many animals exhibit intelligence, rationality, and maturity, language barriers prohibit informed consent.” Footnote 50 Holly Kantin and David Wendler talk of “the lack of a common language,” Footnote 51 and Jane Johnson and Neal Barnard mention “communication barriers” Footnote 52 insofar as animals cannot tell us what kind of research environment is appropriate for them. However, these characterizations are misleading, insofar as the requirement of informed consent can never apply to animals: most animals could never possibly fulfill its ascription-conditions. The concept of language barriers suggests that if animals could talk, then the problem would be resolved. But much more than merely language and communication barriers are at stake when giving informed consent—namely, cognitive capacities such as rationality, the ability to know how to act intentionally in one’s best interest in the long term, understanding complex circumstances, and the like. As Richard Healey and Angie Pepper put it: animals are incapable of giving informed consent because they “cannot understand, form, and communicate complex intentions about normative concepts like rights and duties.” Footnote 53 That is, animals cannot waive rights regarding themselves and their bodies and thus authorize others to undertake an otherwise impermissible action (such as administering a drug). Animals are thus incapable of giving informed consent to their participation in medical research.

Nonetheless, there is another understanding of informed consent that could be applied to the case of animal research, namely assent and dissent. Animals have various preferences, which they can express. For example, animals can show—with the help of humans and a trial-and-error system—their food preferences, or, in the case of dogs, which walking route they wish to take with their human guardian. I argued earlier that animals have a basic right to express normal behavior. Part of animals’ normal behavior is their ability to freely pursue their own interests and preferences. To act upon their will—to move freely, to curiously discover a new area or to freely choose with whom they interact—constitutes a significant part of a good life for many animals. That is, many animals can express their will through their actions and in interactions with other beings—and this capacity should be respected. Applied to the case of animal research, this means that animals can show whether they wish to partake in research or not. Thus, there is a way to reformulate the idea of informed consent in research animals:

The fact that investigators are required to solicit the assent and respect the dissent of human subjects who are unable to provide informed consent suggests that animals’ inability to provide informed consent does not provide a justification for failing to take into account their preferences regarding whether they participate in research. Footnote 54

Kantin and Wendler distinguish welfare- and agency-based reasons for why researchers should respect animals’ preferences to withdraw from or participate in research. Preferences can mirror individuals’ welfare, which provides a pro tanto reason to take them into account. Hence, researchers should observe if animals show dissent (e.g., in the form of discomfort or pain), with the aim of maintaining their welfare. Footnote 55 Although it may be difficult to determine the exact source of unease in research animals (since it may also be due to fear, hunger, and the like), paying attention to dissenting behavioral cues can help to determine whether the harm experienced by the research animals exceeds an allowable upper limit, and corresponding measures could be taken to minimize suffering. That is, respect for sustained dissent can minimize the overall harm incurred by research animals.

Agency-based reasons are concerned with respect for individuals’ capacity to freely choose what they prefer, regardless of whether it is beneficial or detrimental to their well-being. Agency has intrinsic value insofar as it matters for forming and shaping one’s life in the way one wishes. This capacity for agency can sometimes be found in some animals (e.g., great apes, dolphins, and possibly elephants). Kantin and Wendler infer from this fact that, “[…] at a minimum, it seems plausible that investigators have an obligation to make sure that no animal to whom agency-based reasons apply has her dissent disrespected on a consistent basis.” Footnote 56 In practice, this means that different methods or analgesia should perhaps be deployed, or that strongly dissenting animals should be excluded from research.

In some cases, it is also possible to extend the principle of assent to research animals, such as when animals can freely choose to join the laboratory of their own will and when it pleases them, for example in the case of behavioral studies. Under animals’ assent, I understand here animals’ approval of what is happening to them. This can manifest itself in the form of animals’ not showing disapproval or resistance, their showing approval to what is happening to them, or their affirmative behavior, such as when they join a study setting deliberately. Footnote 57 An example are chimpanzees in reserves or sanctuaries who participate in studies involving video games which test their cognitive capacities; they engage in this research of their own will, since it presents them with a welcome distraction from their daily life.

In summary, this means that close attention should be paid to the behavior and preferences of individual animals during studies. Researchers should test whether animals show assent to their study participation, and steps should be taken if dissent is perceived. In practice, this implies that many painful experiments conducted on animals would no longer be possible.

Respect for Research Subjects

As their last requirement, Emanuel, Wendler, and Grady list “respect for research subjects.” The first aspect of this criterion involves permitting withdrawal from the research anytime, a point already covered in the previous section on animals’ assent and dissent. Furthermore, Emanuel, Wendler, and Grady mention the need to protect privacy through confidentiality, to provide information about newly discovered risks or benefits, and to inform research subjects of the clinical research’s results. These requirements do not seem relevant to the case of animals. However, there is a fifth requirement that does matter, namely, that researchers maintain the welfare of their research subjects. This can be described as beneficence. Current research practices often fall short of this requirement. For example, it is a common practice to induce harmful conditions and diseases in animals (such as cancer or lameness)—a practice we should never accept in the case of humans. If animals have a right to remain free of disease and to benefit from bodily integrity, then the same protection should apply to their case: causing them deliberately harmful conditions (such as, e.g., cancer) would be morally impermissible.

This leads us to consider another problematic practice, namely bringing animals into existence merely to serve research purposes. Many animals are bred for research, and they will normally spend their whole lives in research settings, afflicted by disease or other painful conditions while being used in harmful experiments until they die. These animals are not brought into existence for their own sake; rather, they are exclusively bred as means to the end of the research, and usually they will experience a rather low quality of life. If animals have basic rights, such pure instrumentalization is morally problematic. We usually think that it is morally reprehensible to bring children into existence solely as mere means to an end, such as serving their parents. The same should hold for research animals, if they have basic moral rights. As Sue Donaldson and Will Kymlicka note regarding companion animals: “Humans may bring dogs into their lives for pleasure (and company, love, and inspiration), but this is compatible with dogs existing in and for themselves) as it is in the case of humans.” Footnote 58 In practice, humans can still bring domesticated animals into existence, but these animals always have to be valued for themselves: they cannot exclusively be bred for the purpose of animal research alone, being deliberately infected with diseases or having other harmful conditions inflicted upon them. This means that nonharmful or minimally invasive research with already-existing sick animals may be ethically permissible, as long as it is in the interest of these animals to participate in a given study.

Lastly, respect for animals also means that their right to life must be respected: laboratory animals should not automatically be put to death once the study ends, unless unavoidable suffering or pain due to natural causes makes euthanasia necessary. Rather, steps should be taken to allow these animals a fulfilled life in sanctuaries or homes. That is, animals should only be put to death if it is in their own best interest and unavoidable for reducing suffering that cannot be alleviated otherwise. Footnote 59 A possible option is to rehome animals once a study is finished, that is, to find caretakers or sanctuaries for animals used in research Footnote 60 —a practice already suggested for some species by the European Directive on the protection of animals used for scientific purposes. Footnote 61

In this article, I have defended the view that if animals have moral rights, then animal research should not necessarily be abolished, but should rather be completely reformed by modeling it on ethical research with humans. If we entirely excluded animals from research, then we would deprive them of potential benefits to them or their species. Taking the Five Freedoms as a basis, I fleshed out animals’ basic rights and argued that requirements already in place for ethical research with humans should be extended to animal research. If research with humans who cannot speak up for themselves and consent to their participation in studies may be conducted as long as some requirements are met and the risk-level is low, I contended, then the same conditions should apply to animals’ involvement in research. Moreover, I outlined how this can best be done in practice, by assessing how the requirements of social and scientific value, scientific validity, independent review, fair subject-selection, harm–benefit ratio, informed consent, and respect could be applied to animal research.

Given all these considerations, it appears that only noninvasive animal research, some behavioral studies where animals are merely observed, practices involving little or no harm and stress, and the use of the animals after their natural death are ethically permissible. Further cases of admissible research include: “the disease or condition being investigated is one that naturally occurs in the study animal; the animal enrolled in the experiment is already afflicted with that disease or condition; and participation in the research offers the chance of benefit (or no more than minimal risk) to the individual participant.” Footnote 62

Note that the list of rights and research requirements presented here is a rather basic one. That is, the rights and requirements presented form a minimal threshold for ethically acceptable animal research. Other, complementary principles may be added to this list, such as protections for particularly vulnerable groups of animals (e.g., animal species who need special provisions when in research settings due to their specific needs, which may be harder to meet or are more likely to be overlooked). However, the rights and requirements presented here are meant to establish the most basic conditions animal research should meet on the supposition that animals have rights.

To be sure, there may be disagreement about some of the animal rights and requirements I defend in this article. Some people may argue, for example, that the right to continued existence is too demanding, and that animals lack such a right. However, even if I deem the list of requirements presented here jointly necessary for ethical animal research, there remain possible applications of research requirements that are compatible with animals’ deaths. That is, given that we live in a predominantly speciesist world, even the extension of just one of these requirements would already be a step in the right direction. Hence even if one did not accept one or several of the rights or requirements outlined here, then there would nonetheless be other lessons to be learned from the application of the remaining research requirements and rights to the case of animal research.

Acknowledgments

The author wishes to thank the participants of the annual conference of the Chair of Theoretical Philosophy at the University of Basel, as well as the participants of the colloquium in practical philosophy at the University of Berne for their helpful comments on previous versions of this article. This work was supported by the Swiss National Science Foundation (Grant number 179826).

1. Regan , T. The Case for Animal Rights . Berkeley : University of California Press ; 2004 Google Scholar .

2. Cochrane , A. Animal Rights Without Liberation: Applied Ethics and Human Obligations . New York : Columbia University Press ; 2012 Google Scholar .

3. Emanuel , EJ , Wendler , D , Grady , C. What makes clinical research ethical? JAMA 2000 ; 283 ( 20 ): 2701 –11 CrossRef Google Scholar PubMed .

4. Ferdowsian , H , Choe , C. Extending human research protections to non-human animals . In: Corbey , R , Lanjouw , A , eds. The Politics of Species. Reshaping our Relationships with Other Animals . Cambridge : Cambridge University Press ; 2013 , at 232 –40 CrossRef Google Scholar .

5. Kantin , H , Wendler , D. Is there a role for assent or dissent in animal research? Cambridge Quarterly of Healthcare Ethics 2015 ; 24 ( 4 ): 459 –72 CrossRef Google Scholar PubMed .

6. Fenton , A. Can a chimp say “no”? Reenvisioning chimpanzee dissent in harmful research . Cambridge Quarterly of Healthcare Ethics 2014 ; 23 ( 2 ): 130 –9 CrossRef Google Scholar .

7. Arnason , G , Clausen , J. On balance: weighing harms and benefits in fundamental neurological research using nonhuman primates . Medicine, Health Care, and Philosophy 2016 ; 19 ( 2 ): 229 –37 CrossRef Google Scholar PubMed .

8. Ferdowsian , H , Fuentes , A. Harms and deprivation of benefits for nonhuman primates in research . Theoretical Medicine and Bioethics 2014 ; 35 ( 2 ): 143 –56 CrossRef Google Scholar PubMed .

9. Walker , RL . Human and animal subjects of research: The moral significance of respect versus welfare . Theoretical Medicine and Bioethics 2006 ; 27 ( 4 ): 305 –31 CrossRef Google Scholar PubMed .

10. Johnson , J. Vulnerable subjects? The case of nonhuman animals in experimentation . Journal of Bioethical Inquiry 2013 ; 10 : 497 – 504 CrossRef Google Scholar PubMed .

11. Johnson , J , Barnard , ND . Chimpanzees as vulnerable subjects in research . Theoretical Medicine and Bioethics 2014 ; 35 : 133 –41 CrossRef Google Scholar PubMed .

12. Ferdowsian , H , Johnson , LSM , Johnson , J , Fenton , A , Shriver , A , Gluck , J. A Belmont Report for animals? Cambridge Quarterly of Healthcare Ethics 2020 ; 29 ( 1 ): 19 – 37 CrossRef Google Scholar PubMed .

13. Beauchamp , TL , DeGrazia , D. Principles of Animal Research Ethics . New York : Oxford University Press ; 2020 CrossRef Google Scholar .

14. DeGrazia , D , Beauchamp , TL . Beyond the 3 Rs to a more comprehensive framework of principles for animal research ethics . ILAR Journal 2019 ; 59 ( 2 ):il z011 Google Scholar .

15. Similarly, David DeGrazia and Jeff Sebo outline three principles that they deem acceptable for animal research proponents who are sympathetic to the view that animals matter morally, but who think that humans matter more than animals: an expectation of sufficient net benefit; a worthwhile-life condition; and a no-unnecessary-harm/qualified-basic-needs condition ( DeGrazia , D , Sebo , J. Necessary conditions for morally responsible animal research . Cambridge Quarterly of Healthcare Ethics 2015 ; 24 ( 4 ): 420 –30 CrossRef Google Scholar PubMed ).

16. A note about the types of research I am concerned with: my article deals with all types of animal experimentation in controlled settings such as laboratories. Admittedly, biological or ecological field research involving wild animals may be harmful and thus raise ethical issues ( Zemanova , MA . More training in animal ethics needed for European biologists . BioScience 2017 ; 67 ( 3 ): 301 –5 CrossRef Google Scholar ); nonetheless, for present purposes, I am not concerned with the principles that should guide this type of research.

17. Singer , P. Animal Liberation: The Definitive Classic of the Animal Movement . New York : Ecco Book/Harper Perennial ; 2009 Google Scholar .

18. Horta , O. What is speciesism? Journal of Agricultural and Environmental Ethics 2010 ; 23 ( 3 ): 243 –66 CrossRef Google Scholar .

19. Farm Animal Welfare Council (FAWC): Five Freedoms; available at https://webarchive.nationalarchives.gov.uk/20121010012427/http://www.fawc.org.uk/freedoms.htm (last accessed 30 May 2021). Another option to establish a list of animal rights would be to start from the basic rights of humans to see which ones could apply to animals. Assuming that the resulting list would probably be similar to the one I am presenting, I do not pursue this option further.

20. McCulloch , SP . A critique of FAWC’s Five Freedoms as a framework for the analysis of animal welfare . Journal of Agricultural and Environmental Ethics 2013 ; 26 ( 5 ): 959 –75 CrossRef Google Scholar .

21. FAWC. Farm Animal Welfare in Great Britain: Past, Present and Future; 2009; available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/319292/Farm_Animal_Welfare_in_Great_Britain_-_Past__Present_and_Future.pdf .

22. McCausland , C. The Five Freedoms of animal welfare are rights . Journal of Agricultural and Environmental Ethics 2014 ; 27 ( 4 ): 649 –62 CrossRef Google Scholar .

23. McCulloch , SP . On the virtue of solidarity: animal rights, animal welfarism and animals’ rights to wellbeing . Journal of Animal Welfare Law 2012 (Spring/Summer): 5 – 15 Google Scholar .

24. Sapontzis , SF . Morals, Reason, and Animals . Philadelphia : Temple University Press ; 1987 Google Scholar .

25. Rollin , BE . Animal Rights and Human Morality . 3rd ed . Amherst, NY : Prometheus Books ; 2006 Google Scholar .

26. DeGrazia , D. Taking Animals Seriously: Mental Life and Moral Status . Cambridge : Cambridge University Press ; 1996 CrossRef Google Scholar .

27. See note 1 , Regan 2004.

28. See note 3 , Emanuel et al. 2000.

29. Shepherd , V. Research involving adults lacking capacity to consent: the impact of research regulation on ‘evidence biased’ medicine . BMC Medical Ethics 2016 ; 17 ( 1 ): 55 CrossRef Google Scholar PubMed .

30. This is required by the United States Code Animal Welfare Act 2015; available at https://www.govinfo.gov/content/pkg/USCODE-2015-title7/html/USCODE-2015-title7-chap54.htm (last accessed 30 May 2021); Canadian Council on Animal Care research code Guide to the Care and Use of Experimental Animals, Vol. 1; 2020; available at https://www.ccac.ca/Documents/Standards/Guidelines/Experimental_Animals_Vol1.pdf (last accessed 30 May 2021); The Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010 on the protection of animals used for scientific purposes 2010; available at https://eur-lex.europa.eu/eli/dir/2010/63/oj . (last accessed 30 May 2021). One could argue that weighing harms against benefits should also figure on this list of basic research standards, but since it is not an explicit requirement in the Animal Welfare Act, it is not listed here.

31. Baker , D , Lidster , K , Sottomayor , A , Amor , S , Eisen , JA . Two years later: Journals are not yet enforcing the ARRIVE guidelines on reporting standards for pre-clinical animal studies . PLoS Biology 2014 ; 12 ( 1 ):e1001756 CrossRef Google Scholar .

32. Muhlhausler , BS , Bloomfield , FH , Gillman , MW . Whole animal experiments should be more like human randomized controlled trials . PLoS Biology 2013 ; 11 ( 2 ):e1001481 CrossRef Google Scholar PubMed .

33. ter Riet , G , Korevaar , DA , Leenaars , M , Sterk , PJ , Van Noorden , CJF , Bouter , LM , et al. Publication bias in laboratory animal research: A survey on magnitude, drivers, consequences and potential solutions . PLoS ONE 2012 ; 7 ( 9 ): e43404 CrossRef Google Scholar PubMed .

34. Kimmelman , J , Anderson , JA . Should preclinical studies be registered? Nature Biotechnology 2012 ; 30 ( 6 ): 488 –9 CrossRef Google Scholar PubMed .

35. Jansen of Lorkeers , SJ , Doevendans , PA , Chamuleau , SAJ . All preclinical trials should be registered in advance in an online registry . European Journal of Clinical Investigation 2014 ; 44 ( 9 ): 891 –2 CrossRef Google Scholar .

36. Pound , P , Bracken , MB . Is animal research sufficiently evidence based to be a cornerstone of biomedical research? British Medical Journal 2014 ; 348 : g3387 CrossRef Google Scholar PubMed .

37. Begley , CG , Ioannidis , JPA . Reproducibility in science: Improving the standard for basic and preclinical research . Circulation Research 2015 ; 116 ( 1 ): 116 –26 CrossRef Google Scholar PubMed .

38. Jarvis , MF , Williams , M. Irreproducibility in preclinical biomedical research: Perceptions, uncertainties, and knowledge gaps . Trends in Pharmacological Sciences 2016 ; 37 ( 4 ): 290 – 302 CrossRef Google Scholar PubMed .

39. Begley , CG , Ellis , LM . Drug development: Raise standards for preclinical cancer research . Nature 2012 ; 483 ( 7391 ): 531 –3 CrossRef Google Scholar PubMed .

40. Kilkenny , C , Browne , WJ , Cuthill , IC , Emerson , M , Altman , DG . Improving bioscience research reporting: The ARRIVE guidelines for reporting animal research . PLoS Biology 2010 ; 8 ( 6 ):e1000412 CrossRef Google Scholar PubMed .

41. Hansen , LA . Institution animal care and use committees need greater ethical diversity . Journal of Medical Ethics 2013 ; 39 ( 3 ): 188 –90 CrossRef Google Scholar PubMed .

42. Plous , S , Herzog , H. Animal research: Reliability of protocol reviews for animal research . Science 2001 ; 293 ( 5530 ): 608 –9 CrossRef Google Scholar PubMed .

43. Vogt , L , Reichlin , TS , Nathues , C , Würbel , H. Authorization of animal experiments is based on confidence rather than evidence of scientific rigor . PLoS Biology 2016 ; 14 ( 12 ):e2000598 CrossRef Google Scholar PubMed .

44. Nobis N. Why IACUCs need ethicists. ILAR Journal 2020; ilaa021.

45. Choe Smith , CU . Confronting ethical permissibility in animal research: Rejecting a common assumption and extending a principle of justice . Theoretical Medicine and Bioethics 2014 ; 35 : 175 –85 CrossRef Google Scholar PubMed .

46. Kopelman , LM . Minimal risk as an international ethical standard in research . The Journal of Medicine and Philosophy 2004 ; 29 ( 3 ): 351 –78 CrossRef Google Scholar PubMed .

47. Binik , A. On the minimal risk threshold in research with children . The American Journal of Bioethics 2014 ; 14 ( 9 ): 3 – 12 CrossRef Google Scholar PubMed .

48. Beauchamp , TL , Morton , DB . The upper limits of pain and suffering in animal research . Cambridge Quarterly of Healthcare Ethics 2015 ; 24 ( 4 ): 431 –47 CrossRef Google Scholar PubMed .

49. See note 30 , Directive 2010/63/EU.

50. See note 4 , Ferdowsian, Choe 2013, at 232–40.

51. See note 5 , Kantin, Wendler 2015, at 459–72.

52. See note 11 , Johnson, Barnard 2014, at 133–41.

53. Healey , R , Pepper , A. Interspecies justice: Agency, self-determination, and assent . Philosophical Studies 2020 ; 178 : 1223 –43 CrossRef Google Scholar .

54. See note 5 , Kantin, Wendler 2015, at 459–72.

55. Andrew Fenton made a similar proposition regarding dissent in research animals in general (Fenton A. Holding animal-based research to our highest ethical standards: Re-seeing two emergent laboratory practices and the ethical significance of research animal dissent. ILAR Journal 2020, ilaa014), and chimpanzees in particular (see note 6 , Fenton 2014, at 130–9).

56. See note 5 , Kantin, Wendler 2015, at 459–72.

57. For a detailed account of animal’s assent and the conditions it needs to fulfill, see note 53 , Healey, Pepper 2020. Gardar Arnason recently argued that animals are incapable of assenting because assent presupposes understanding information about what is going to happen during an event or procedure ( Arnason , G. The emergence and development of animal research ethics: A review with a focus on nonhuman primates . Science and Engineering Ethics 2020 ; 26 ( 4 ): 2277 –93 CrossRef Google Scholar PubMed ). However, according to his view, animals can show acquiescence to what is happening to them. Note, although, that Arnason refers to a more demanding account of assent than Healey and Pepper, whose conception I am following here in broad terms.

58. Donaldson , S , Kymlicka , W. Zoopolis: A Political Theory of Animal Rights . Oxford : Oxford University Press ; 2011 Google Scholar .

59. Similarly to humans who should have access to assisted dying if they deem their suffering as too unbearable.

60. See note 55 , Fenton 2020.

61. See note 30 , Directive 2010/63/EU.

62. See note 11 , Johnson, Barnard 2014, at 133–41.

This article has been cited by the following publications. This list is generated based on data provided by Crossref .

• Google Scholar

View all Google Scholar citations for this article.

Save article to Kindle

To save this article to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle .

Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

• Volume 31, Issue 1
• Angela K. Martin (a1)
• DOI: https://doi.org/10.1017/S0963180121000499

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox .

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive .

Reply to: Submit a response

- No HTML tags allowed - Web page URLs will display as text only - Lines and paragraphs break automatically - Attachments, images or tables are not permitted

Your details

Your email address will be used in order to notify you when your comment has been reviewed by the moderator and in case the author(s) of the article or the moderator need to contact you directly.

You have entered the maximum number of contributors

Conflicting interests.

Please list any fees and grants from, employment by, consultancy for, shared ownership in or any close relationship with, at any time over the preceding 36 months, any organisation whose interests may be affected by the publication of the response. Please also list any non-financial associations or interests (personal, professional, political, institutional, religious or other) that a reasonable reader would want to know about in relation to the submitted work. This pertains to all the authors of the piece, their spouses or partners.

• University News
• Faculty & Research
• Health & Medicine
• Science & Technology
• Social Sciences
• Humanities & Arts
• Students & Alumni
• Arts & Culture
• Sports & Athletics
• The Professions
• International
• New England Guide

The Magazine

• Current Issue
• Past Issues

Class Notes & Obituaries

• Browse Class Notes
• Browse Obituaries

Collections

• Commencement
• The Context
• Harvard Squared
• Harvard in the Headlines

Support Harvard Magazine

• Why We Need Your Support
• How We Are Funded
• Ways to Support the Magazine
• Special Gifts
• Behind the Scenes

Classifieds

• Vacation Rentals & Travel
• Real Estate
• Products & Services
• Harvard Authors’ Bookshelf
• Education & Enrichment Resource
• Ad Prices & Information
• Place An Ad

Follow Harvard Magazine:

Social Sciences | 7.25.2019

From the Archives: Animal Research

Every year, scientists use millions of animals—mostly mice and rats—in experiments. the practice provokes passionate debates over the morality and efficacy of such research—and how to make it more humane..

Click image to see full cover: The original January-February 1999 issue that this article appeared in

Read the original article as it appeared in 1999.

The volume of biomedical research, and of trials of new therapies, has increased dramatically in recent decades, fueled by advances in understanding of the genome and how to manipulate it, methods of processing huge data sets, and fundamental discoveries such as targeted and immunological approaches to attacking cancer (see “Targeting Cancer,” May-June 2018). Greater Boston, and Harvard, are major participants in academic biomedical research, in close proximity to the biotech and pharmaceutical industries, which have set up shop locally to tap into the wealth of talent and ideas. Along the way, new research techniques that are driven by more sophisticated imaging, bioinformatics (see “Toward Precision Medicine,” May-June 2015), and “organ-on-a-chip” technology have made it possible to conduct science with less reliance on various kinds of animal-testing. Given rising social concern for and interest in animal welfare (see “Are Animals ‘Things’?” March-April 2016), these converging trends make rereading this in-depth 1999 report by John F. Lauerman on the use of animals in biomedical research still timely and important. ~The Editors

“What is man without the beasts? If the beasts were gone, man would die from a great loneliness of spirit. For whatever happens to the beasts soon happens to man.”

~ Chief Seattle

Frederick Banting would never have begun his research without access to research animals. Before he had even spoken of his ideas, his first note to himself on the subject read: "Ligate the pancreatic ducts of dogs." The quiet Ontario doctor envisioned that severing the connection between the pancreas and the digestive system in a living animal would allow him to isolate the mysterious substance that would control diabetes.

During the first week in the laboratory, Banting and his assistant, Charles Best, operated on 10 dogs; all 10 died. Finally, in 1921, after months of experimentation, Banting and his colleagues isolated a material that kept a depancreatized dog named Marjorie alive for about 70 days. Exactly what information was gained from using dogs, and how many dogs were absolutely needed, is not clear. Work previous to Banting and Best’s, some of it in humans, had indicated the presence and importance of a hormone involved in glucose transport. Many more experienced scientists in the diabetes-research community believed that Marjorie had never been fully depancreatized, and thus may have never been diabetic. More likely, they said, the dog died of infection caused by her pancreatectomy. It’s possible that even the death of the famous Marjorie was unnecessary for the great discovery.

But the two Toronto researchers had isolated insulin, providing the first step toward producing it from pig and cow pancreas, available in bulk from slaughterhouses. The result—that Banting and Best "saw insulin"—appears to have justified all sacrifices. What’s the life of a dog, 10 dogs, a hundred? Before Banting and Best operated on dogs, we had no insulin; afterwards, we did.

Stories such as these are the reason our society and the vast majority of societies in the world accept the use of animals as a vital component of medical research.

Deeply entrenched traditions support the notion that animal welfare must bow to the best interests of humans. Animal domestication was among the first labor-saving devices. Humans have experimented with animal breeding, feeding, and disease control for thousands of years—not to benefit the animals themselves, but to insure that the owners obtained a maximum yield.

Today, those traditional practices have evolved into a scientific institution, the appropriateness of which is subject to perennial debate. In the United States alone, there are an estimated 17 million to 22 million animals in laboratory research facilities. To many people, animal research represents a doorway to the medical treatment of tomorrow. But to animal protectionists, and a growing number of other Americans, animal experimentation is a barbaric, outdated practice that—on the basis of a few notable past successes—has somehow retained its vestigial acceptability.

"Let’s say that it’s true, that animals were indispensable to the discovery of insulin," says Neal Barnard, M.D., of the Physicians Committee for Responsible Medicine, an animal-protection group. "That was a long time ago. I think to say, ‘It was done this way and there’s no other way it could have been done’ is a bit of a leap of faith, but let’s say that at the time there was no other way. You could also say that you couldn’t have settled the South without slavery. Would you still do it that way today? Just because something seemed necessary or acceptable at the time is not to say that we should do it in our time."

The Animal Debate

The legitimation of the animal-research debate challenges one of the most important and widely used scientific approaches to discovery about the human body and its diseases. Animal experimentation is often considered as much of a sine qua non to research as the Bunsen burner. But animal protectionists reply that the importance of animals to research is overrated, and that their pressure has exposed profligacy among experimenters.

In February 1997, a highly controversial collection of articles appeared in  Scientific American  on the subject of laboratory-animal research. The first, written by Barnard and Stephen Kaufman, M.D., of the Medical Research Modernization Committee, another protectionist group, advanced the view that data collected from animal experimentation are almost always redundant and unnecessary, frequently misleading, and by their very nature unlikely to provide reliable information about humans and their diseases. "Animal ‘models’ are, at best, analagous to human conditions," the authors wrote, "but no theory can be refuted or proved by analogy. Thus, it makes no logical sense to test a theory about humans using animals."

A rebuttal in support of animal research followed, by Jack Botting, Ph.D., former scientific adviser to the Research Defense Society in London, and Adrian Morrison, Ph.D., D.V.M., of the University of Pennsylvania School of Veterinary Medicine. Their reply cited examples of scientists from Louis Pasteur to John Gibbon, a twentieth-century pioneer in open-heart surgery, who made important breakthroughs in the treatment of human disease through animal research.

Many scientists—both supporters of animal research and advocates for its diminution—simply refused to discuss the difficult topic, recalls Madhusree Mukerjee, the editor who proposed that  Scientific American  explore the controversy and who wrote a third article, reporting on the overall state of animal research in the sciences. (Similar difficulties were encountered in researching the present article.) Mukerjee suspects that possible interviewees feared the criticism of their colleagues.

Reader response, on the other hand, was overwhelming, both pro and con. "We got a huge amount of flak for dealing with the subject at all," recalls Mukerjee. "Some of it was fairly frightening." To many animal-research supporters, it was as though the floodgates had been opened. "I am simply stunned that  Scientific American,  a paragon of promotion of scientific research, would actually offer up for debate whether animal research should occur," wrote one reader. "Please leave this question of animal research to animal-rights activists, and stop yourselves from turning into scientific wimps." "A lot of the scientific community felt [ Scientific American’ s editors] had overstepped their bounds and compromised their values by printing the Barnard-Kaufman article," says Joanne Zurlo, associate director of the Johns Hopkins School of Public Health Center for Alternatives to Animal Testing and a specialist in chemical carcinogenesis.

Those researchers who supported animal use and wrote in said the animal-protectionists’ side of the  Scientific American  debate was fraught with misstatements and scientific errors, although Mukerjee maintains that all the articles were painstakingly fact-checked. "We annoyed a lot of influential scientists," she says. "Our publication has spent more than a century describing advances in medical research, including some by fairly controversial figures. We’d never addressed the question of research on animals before, and in a sense it was a necessary thing to do. We probably lost some subscriptions because of it. But we are a bridge between the researchers who write for us and the public who read us, and we decided to let our readers decide for themselves."

Animal Welfare

Animal protectionists date their movement back to the times of Leonardo da Vinci and even Pythagoras, who are alleged to have been vegetarians. Numerous essayists and animal lovers have detailed their objections to the misuse of animals. Yet not long ago, virtually anyone who wanted to could conduct experiments on animals. In the 1960s, it was not uncommon to walk into a laboratory and find mice, dogs, cats, even monkeys, housed on the premises in whatever conditions researchers saw fit to provide. Banting himself frequently bought pound dogs and may even have caught dogs on his own; his collaborators recalled that he once arrived at the lab with a dog he had leashed with his tie.

Only in the nineteenth century did animal research begin to draw explicit objections from protectionists. A pivotal event occurred in England in 1874, when a lecturer at the University of Norwich demonstrated how to induce epileptic symptoms in a dog through the administration of absinthe. Objections were raised by students in the audience, and the dog was set free. Later, charges were filed against the lecturer under Dick Martin’s Act, an 1822 law that called for a fine of 10 shillings from anyone committing acts of cruelty against animals. Two years later, in 1876, Parliament passed the Cruelty to Animals Act, requiring a license for animal experimentation and placing restrictions on some painful forms of experimentation.

In the United States, minimal restrictions on animal experimentation prevailed until 1966, when the first federal Laboratory Animal Welfare Act (now known as the Animal Welfare Act, or AWA) was passed by Congress. In 1970 the AWA was broadened to require the use of appropriate pain-relieving drugs, and to include commercially bred and exhibited animals. Six years later, provisions were added covering animal transport and prohibiting animal-fighting contests. In 1985, Congress passed the Improved Standards for Laboratory Animals Act, which again strengthened the AWA by providing laboratory-animal-care standards, enforced by U.S. Department of Agriculture (USDA) inspectors, and also aimed to reduce unnecessarily duplicative animal-research experimentation.

In 1976, however, the AWA was amended in a rather curious way: rats, mice, birds, horses, and farm animals were specifically excluded from its purview for reasons that are not fully clear, although the USDA’s limited resources—along with political pressure from interested parties—are likely to be among them. Since rats and mice make up more than 95 percent of all research animals in this country, the amendment effectively put the vast majority of laboratory animals outside the reach of the USDA. Since then, at least one court has ruled the 1976 amendment "arbitrary and capricious."

The Mouse Warehouse

As associate professor of surgery Arthur Lage, D.V.M., walks through the doors of Harvard Medical School’s Alpert Building, people recognize him, smile, and let us pass without showing identification. He is director of the Center for Animal Resources and Comparative Medicine and the Center for Minimally Invasive Surgery at the medical school and director of the Office of Animal Resources for the Faculty of Arts and Sciences as well. We take an elevator down to a basement, where Lage swipes a card through a reader, unlocking a door to a hallway, where he speaks into a phone. A minute later, a young man clad in blue scrubs opens the door. Lage explains that he’s bringing a reporter in for a tour and that we’ll need keys to see certain rooms. The young man hands over the keys and closes the door.

At the other end of the short hallway are two doors, each leading to a sanitary changing room. When you turn the lights on in the changing rooms, the doors at either end lock automatically. After we’ve pulled blue scrubs over our clothes, Lage douses the lights and we step out of the room into another brightly lit hallway.

We’re in one of Harvard’s 16 animal facilities now, a moderately "clean" facility—meaning that it requires only minimal preparations for entry. Some laboratories would require us to remove our clothes and shower before entering; others don’t even stock scrubs. But this facility is full of mice—transgenic mice. A stray pathogen in one of the animal rooms could wipe out millions of dollars’ worth of experiments or, just as disastrous, infect a colony of mice with viruses or bacteria that might confound the results of a study.

Of course, the security isn’t intended only to repel microbes. Perhaps in frustration with perceived shortcomings in the oversight of animal experimentation, some animal-protection groups have gained a reputation for tactics that are rash and often destructive. On several occasions, animals have been "liberated" from laboratories, erasing potential results and sometimes careers. In 1989, the Animal Liberation Front took credit for the release of more than 1,200 laboratory animals, some of them infected with cryptosporidium, which can be harmful to infants and immunocompromised people. The total damage was estimated at $250,000. In 1987, a laboratory under construction at the University of California at Davis was burned; the loss was estimated at $3 million.

Although there is little evidence of violence toward animal researchers here in the United States, in Europe, where the animal- protection movement is more firmly entrenched, activists have taken aim at individuals, sometimes with disastrous results. In 1990, the infant daughter of a researcher was injured by a car bomb believed to have been set by animal protectionists. In separate, related incidents, a furrier and a breeder of cats used in experimentation were injured by letter bombs. Responsibility for the mail bombs was assumed by "The Justice Department," a militant, underground, animal-protection organization.

Even today, animal-protection groups find ways to gain access to research and testing facilities. In 1997, Michelle Rokke of People for the Ethical Treatment of Animals (PETA) infiltrated Huntingdon Life Sciences, a drug- and cosmetic-testing firm in East Millstone, New Jersey. Using a surveillance camera embedded in her eyeglasses, Rokke took hours of films that PETA claimed showed animals being slammed into cages and roughly handled. PETA president and co-founder Ingrid Newkirk said their investigation also revealed that young beagles’ legs were broken for another study at Huntingdon. Movie star Kim Basinger gave a press conference on Huntingdon’s lawn. In April 1998, the USDA fined Huntingdon $50,000 for AWA violations.

In the basement of the Alpert building, there is no evidence of such fury. Each room holds literally hundreds of mice in shoebox-sized cages, and there are so many of them it looks like a shoe warehouse. There are about 55,000 mice involved in research at Harvard at any one time, but that number is growing constantly. In 1997 it was closer to 50,000; by the end of 1998 it approached 58,000. By comparison, the numbers of other animals are almost negligible: about 1,300 rats, 145 rabbits, 115 hamsters, 70 guinea pigs, 67 primates, 35 pigs, 30 gerbils, 25 chicks, 20 dogs, 18 sheep, 6 cats, and 1 ferret. In addition, the New England Regional Primate Research Center in Southborough, Massachusetts, houses another 1,500 monkeys and other primates. Established at Harvard in 1966 with a grant from the National Institutes of Health, the NERPRC is one of seven such centers created by Congress in the early 1960s to serve as regional resources for scientists.

Surprisingly, there is no hint of animal smell within the basement facility. Temple Grandin of Colorado State University, a specialist in the behavior of captive animals, says that what mice really crave is some form of bedding—wood chips, paper, or shavings—which not all these animals have. Still, these laboratory animals, born and bred under fluorescent lights, are comfortable enough to live out lifespans they would never approach in the wild and, of course, to reproduce. And since almost all of them are involved in genetic studies, making sure they’re happy and healthy enough to reproduce is of vital importance. Keeping these buildings clean and free of infection is a triumph of research design. All the soiled animal cages are shuttled to one end of the laboratory where, before they re-enter, they pass through an enormous autoclaving machine that sanitizes the cages as well as the carts they sit on.

Amid the towers and technology of the medical area, animals one normally associates with a farm are a jarring sight. But Lage (pronounced lah-gee) led me through animal laboratories in the basement of the Seeley Mudd Building where we saw pigs, sheep, and rabbits held in small, clean pens. At one point, we watched eight sheep slated for experimental surgery frisk around a room that looked almost exactly like an office. If the straw were swept away, one could easily have moved in a desk and gone to work.

"We care for all these animals just as though they were covered by the [Animal Welfare] Act," Lage says proudly. "I think most of us believe that the act should cover rats and mice."

Although the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC), like the USDA, inspects laboratory-animal facilities, including those of rats and mice, AAALAC accreditation isn’t legally required to conduct animal research. "AAALAC conducts something like a ‘peer review’ assessment," Lage says. "It’s a voluntary process, subscribed to by many, many research organizations. If you decide not to go through accreditation, you have to describe your entire program every time you apply to the government for funding for animal research."

Many laboratories and commercial drug-testing companies that receive no funding from federal sources and use only rats and mice proceed with only minimal oversight from their own institutional animal care and use committees (IACUCs). But restrictions on animal research are, if anything, increasing, not abating. Not content with the level of state and federal regulation, for example, the city of Cambridge in 1989 passed its own law creating an inspector’s office with the power to make USDA-type inspections of all research facilities housing vertebrate animals, including rats and mice.

Cambridge’s current commissioner of laboratory animals, Julie Medley, D.V.M., annually inspects 34 laboratories, makes follow-up visits to some facilities (sometimes unannounced), and reviews "hundreds and hundreds" of research protocols to ensure that all experiments meet federal standards for pain control. Investigators readily comply with Medley’s suggestions for better animal care and pain control, she says, but she perceives an undercurrent among some researchers who chafe under what they perceive as excessive government intervention in their work. "I’m sure some of the principal investigators resent these regulations," she says. "It doesn’t happen that often, but there are rare occasions when I run into resistance from an investigator."

Still, for animal protectionists, the intentions of the Animal Welfare Act, AAALAC, and state inspectors are not enough. Sandi Larson, a scientific adviser to the New England Anti-Vivisection Society, who has a master’s degree in microbiology, concedes that "not all researchers are Dr. Frankensteins. But," she adds, "they have been trained to look at animals as tools. It’s ingrained in them to shut off their compassion and act like scientists. They think there’s no room for emotions." A significant portion of the animal-protection movement believes that most experimentation on animals is without merit. If animals are different enough from humans that we can dismiss their suffering as inconsequential, isn’t it just a little too convenient that they resemble us enough to be considered a source of reliable information about human physiology?

Animal Liberation 

Peter Singer was an Oxford philosophy student who had little interest in animals, domesticated or otherwise, until he had lunch with a vegetarian friend one day and they began talking about the use and abuse of animals. Singer was quickly converted to the cause, and within a few years became its champion. One of the pivotal events in the treatment of laboratory animals in this country and throughout the world was the publication of his manifesto,  Animal Liberation,  in 1975.

Just 25 years ago, some proponents of animal experimentation still held that animals’ intellectual inferiority to humans meant that they could not be accorded the same rights as humans. Some argued that animals had no consciousness or memory, that they did not think as humans did. The quality and intensity of the pain felt by animals was still subject to debate. Singer, recently appointed DeCamp professor of bioethics at Princeton University’s Center for Human Values, refuted the assertion of animals’ inequality, pointing out that our society grants equal rights to all humans without regard to IQ or ability to function. "If the demand for equality were based on the actual equality of all human beings, we would have to stop demanding equality," he wrote. "...[T]he claim to equality does not depend on intelligence, moral capacity, physical strength, or similar matters of fact."

As for consciousness and the ability to feel pain, Singer pointed out that we have no reason to believe animals lack either one. Some of the experiments he recounts make their emotional vulnerability all too clear. In the late 1950s, for instance, psychologist Harry Harlow of the University of Wisconsin embarked on a series of experiments in which he deprived young rhesus monkeys of contact with their mothers. Young monkeys who were most completely deprived of parental contact developed very bizarre behavior, and would cling to objects that supplied the most minimal comfort, such as a scrap of terrycloth. Many of his fellow researchers considered Harlow a genius for having established the importance of interpersonal contact to normal childhood development. Singer, on the other hand, pointed out that the experiments demonstrated just how much like us monkeys really are, and he condemned the inhumanity of torturing them to obtain information that could have been elucidated in many other ways, perhaps through epidemiological studies of children who had been separated from their mothers at critical periods of development.

"You can’t have it both ways," says biochemist Karin Zupko ’77, an animal-rights advocate formerly with the New England Anti-Vivisection Society. "You can’t say that animals are different enough from people so that it’s acceptable to experiment on them, but enough like people so that the results of the experiments are valid."

Models for Medicine

Scientists, however, counter that you can, in fact, gather useful information about humans from animals that seem vastly different from us. They point to the many surgical experiments performed on pigs, dogs, and monkeys that have led to advances in transplantation, heart-valve replacement, and coronary artery bypass graft surgery.

"Research on live organisms is essential for medical advance," asserts Francis D. Moore ’35, M.D. ’39, S.D. ’82, Moseley professor and surgeon-in-chief emeritus at Harvard Medical School and Brigham and Women’s Hospital, respectively. As Moore has pointed out in testimony to the Massachusetts legislature and in his autobiographical book,  A Miracle and A Privilege,  the first successful human kidney transplant, in which Moore played a pivotal role in 1963, would not have been possible at that time without an understanding of immunology based on experiments in rats and mice. Important aspects of the surgery were developed in larger animals. "There’s no substitute for it," says Moore. "Some people say you can set up a computer program to act like a dog. Well, forget it. All animals have responses that we don’t understand, and there’s no way to set that up on a computer."

A great deal of our understanding of basic human physiology comes from experiments in large animals, like dogs and chimpanzees. Harvard physiologist Walter B. Cannon, A.B. 1896, M.D. ’00, S.D. ’37, for example, performed experiments on dogs for many years to understand the basic dynamics of digestion. Different animals may be selected for different purposes. A dog’s prostate differs from that of a human in having only two lobes, yet dogs, like humans, can develop benign prostatic hyperplasia.

"Not all animal models are ideal, but some cases are a perfect fit," says Arthur Lage. "Mice are certainly a very good model for studying human genes. Much of the genetic makeup of the mouse is very similar to that of a human; there are large regions of shared identity." That’s why, Lage explains, Harvard will probably double its use of mice over the next five years—to about 100,000 mice annually. The chief reason for this is transgenic-mouse technology—which allows the insertion and deletion of key disease genes into the mouse genome. These techniques allow researchers to study the impact of both subtle and drastic changes in the genome, and to make key predictions about how similar changes would affect humans. Mice can be bred, for example, with varying ability to express the  p53  gene, which has been implicated in a wide variety of cancers. Understanding how the activity of such genes affects cancer development promises to vastly increase our knowledge of treatment and prevention.

Philip Leder ’56, M.D. ’60, Andrus professor of genetics and head of the medical school’s department of genetics, who pioneered the technology, points out that transgenic mice have been used to test the safety and efficacy of new therapeutics; to detect biohazards; and to advance our knowledge of cancer. Yet he concedes this widely embraced methodology has yet to produce new therapies itself. "It’s impossible as yet to bring it home to lives of patients," he says, "because the development of diagnostics and therapeutics takes time."

There are many areas, however, where a direct connection between animal research and patient welfare can be argued. In the field of AIDS, for instance, research on animals has been making important contributions to the basic understanding, prevention, and treatment of this life-threatening disease.

In 1981, Norman Letvin ’71, M.D. ’75, received a call that would change his life. It concerned an epidemic of mysterious deaths, all caused by unusual pathogens and cancers, such as pneumocystis carinii pneumonia, cytomegalovirus, and rare lymphomas. But the patients suffering from these infections were not humans, but laboratory monkeys.

We now recognize these so-called "opportunistic infections" as signals of the presence of the human immunodeficiency virus (HIV) that causes AIDS. But at that time, the disease was just being recognized in humans, the term "AIDS" itself was unknown, and the cause of all these infections was still a frightening mystery.

Letvin, now professor of medicine at Harvard, says HIV probably began as a relatively harmless virus that infected some species of African monkeys. When it crossed species lines, it did so in several directions, spreading simultaneously into both human and additional non-human primate populations. In these new populations, the infection had much more serious consequences than in the African monkeys: it was lethal. But to Letvin, the realization that a parallel syndrome was occurring in man and monkeys was a tremendous opportunity.

"A great deal of effort has been expended on trying to find rodent and rabbit models for studying HIV infections, but they have not proven terribly useful," Letvin notes. "The only way we can see what happens in the first few minutes, hours, and days after infections—questions that are essential to answer in order to develop an HIV vaccine—is by working in animal models. We are forced to work in these models if we want to answer these questions." (The number of monkeys needed for such an experiment, he hastens to point out, is relatively small: usually about six.)

In Letvin’s experiments, monkeys are inoculated with candidate vaccines against HIV. After a brief period during which the vaccine draws a response from the host monkeys’ immune system, the animals are inoculated with a strain of immunodeficiency virus that brings on an AIDS-like disease. Periodic blood samples are taken to monitor their white blood cell counts and viral replication. An experimental model that causes the monkeys to get sick is more informative, Letvin explains, because even if the vaccine doesn’t prevent infection, it may slow the course of the disease enough to be useful.

"There’s little question that exciting animal data is a major drive for the initiation of human studies," Letvin says. "It’s not a gatekeeper, but an important piece of a complex puzzle we use to determine whether to go forward with the long march into humans. There are hundreds of approaches one could take. If a strategy does look promising, an animal trial makes it easier to determine whether it’s worth spending millions of dollars to measure its safety and efficacy in humans."

Letvin points out that an AIDS vaccine would save millions of human lives, particularly in populations where expensive treatment is not available. Thus the use of animals in research on diseases such as AIDS seems fated to continue for years to come. If the past is any indication, it will probably yield a rich crop of new medical information.

Perhaps the more accurate question then—under the circumstances—is, how much do we care about animal suffering? Is it worthwhile to consider that issue in our quest for better treatment for diseases?

The Three R’s

Since Peter Singer formulated his ideas, the animal-protection movement has gone from a series of staccato eruptions to a steady influence on the course of medical research. Everyone involved in the animal-research debate admits that the situation has changed considerably during the last 25 years. Ernie Prentice, a nationally recognized expert in the regulation and ethics of animal research and a member of the institutional animal care and use committee at the University of Nebraska Medical Center, can remember a time when animals were routinely subjected to painful measures without pain control. In one well-publicized experiment, pigs were burned without anesthetic; in another long-running research project, monkeys were subjected to traumatic blows to the head without analgesics. Animals progressed to the end stages of artificially induced malignancies, renal failure, and heart disease, all without any form of pain control.

"Those kinds of projects would not be permitted now. They would be unacceptable for at least two reasons," says Prentice. "One is that we now have regulations that clearly ban this kind of experimentation, and those regulations are adequately enforced to make sure that they’re followed. At the same time, there is heightened ethical sensitivity among both researchers and IACUCs. If you had sat in on a meeting of an IACUC in 1985 and were able to compare the level of discussion back then with what goes on today, you would see a tremendous difference."

Increasingly, members of the protection community are taking legal steps to gain input into animal-treatment guidelines, and have found more conventional ways to exert pressure. Marc Jurnove, a member of the Animal Legal Defense Fund (ALDF), is suing the USDA for "aesthetic and recreational injuries" that he suffered when seeing the living conditions of chimpanzees and apes at a Long Island zoo. Jurnove charged that the USDA failed to adopt and enforce adequate standards for the animals’ well-being, as is required by the AWA. This past September, the U.S. Court of Appeals for the District of Columbia Circuit, the nation’s most influential circuit court, upheld Jurnove’s right to sue. Recently, the ALDF also led animal-rights groups in successfully suing the National Academy of Sciences for access to records and to committee meetings pertaining to a guide on the care and use of laboratory animals.

Some major funding organizations have also embraced the animal-rights movement. The Doris Duke Charitable Foundation, with assets of $1.25 billion, is one of the 25 wealthiest philanthropies in the country. Although it funds medical research, one of its restrictions is that animals not be used as subjects. This creates a sticky situation for the board, which hopes to fund research on AIDS, cancer, heart disease, and sickle-cell anemia, areas heavily dependent on animal research in the past.

But the effort to occupy a middle ground, supporting the principles of reduction, replacement, and refinement of animal research while acknowledging its necessity, has been extremely frustrating.

Several research institutions have established centers of animal-rights advocacy. The Center for Animals and Public Policy at Tufts University and the Center for Alternatives to Animal Testing at Johns Hopkins University, for example, have tried to establish liaisons with both protectionists and researchers. "I wasn’t running around throwing bombs," says Andrew Rowan, Ph.D., former director of the Tufts Center and now senior vice president of the Humane Society of the United States. "I was engaging colleagues in scientific debate without being obstreperous. People were shouting past each other." Veterinarian Peter Theran, vice president of the health and hospitals division of the Massachusetts Society for the Prevention of Cruelty to Animals and director of the MSPCA’s Center for Laboratory Animal Welfare, says that his group has had to walk a fine line. "We try to maintain a rapport with both sides," he stresses. "I have to say that we often don’t agree with some of the more aggressive groups, like PETA. But there’s a tendency to paint the animal-welfare community with a broad brush. And that makes dialogue extremely difficult."

"When you say you’re for animal welfare, you’re perceived as rabid," says Joanne Zurlo of the Johns Hopkins center. "At the same time, we can’t deal with groups like PETA because they believe in abolition of animal use. When we organized the first World Congress on Alternatives and Animal Use in the Life Sciences in 1993, we invited representatives from every organization to sit at the table. PETA would not join. Even the American AntiVivisection Society sent a representative, but members of the hard-line groups who were picketing outside hounded her and called her a murderer."

Human Lives, Humane Experiments

The growth of the animal-protection debate has been fraught with acrimony. The results, however, go beyond the additional credibility that has been afforded animal protectionists. Scientists, too, find that they can be more open about the feelings they have or may have had for the creatures in their care, and are more free to explore alternative methods of experimentation.

"All of us, whether we’re doing research on animals or not, recognize that this is something that is not optimal," Andrew Rowan says. "If society didn’t feel that we needed the information, we wouldn’t do research on animals. But society feels we do, and so do scientists. There’s a tension between our concern about causing pain and distress and killing animals and our need for new knowledge. No one would say that the animals in research benefit from it, and in a world that was perfect we wouldn’t be doing this. We’re engaged in encouraging people to make animal welfare a higher priority without compromising their ability to gather information."

Neal Barnard, of the Physicians Committee for Responsible Medicine, argues that the route away from animal research should carry us toward population-based efforts like the Framingham Heart Study, in which heart researchers have closely followed the health habits and outcomes of 5,000 adults for just over 50 years. That study was a key factor in galvanizing current national efforts to lower cholesterol, combat hypertension, and encourage proper diet and exercise to reduce mortality from heart disease.

"Those areas where we struggle the most, clinically, are those where we haven’t exploited good clinical research and are relying on animal models," Barnard says. "Look at cardiac defects. We don’t know how they’re caused because no one has done the equivalent of the Framingham study for heart defects, even though it’s quite feasible. The Centers for Disease Control and organizations study these congenital abnormalities only in a very haphazard way.

"Of course," he continues, "there have been some brilliant exceptions, such as the research on neural tube defects. It was found through observation of humans that these defects were associated with deficiencies in folic acid, and that by taking vitamin supplements you can reduce the risk. The same with fetal alcohol syndrome: the breakthroughs came in studying humans, not animals."

Politics frequently obscures our view of research bias, Barnard says. He has called for a Framingham-style study of the health implications of cow’s milk consumption, which has been implicated in some studies as a possible cause of Type 1 diabetes in children. Barnard believes that the political strength of the dairy industry has kept such a study from becoming a reality even though some 700,000 Americans suffer from Type 1 diabetes.

Even within the scientific community, there is an increasing willingness to admit that current research methods can be improved upon. A wide variety of in vitro tests have been proposed (among them, the use of human tissue culture and in vitro cell-culture assays), as well as increased reliance on computer modeling and the creative application of human epidemiological studies. Both government and industry experts agree that if new techniques eliminate or reduce the use of animals, so much the better. "[T]he current rodent bioassay for assessing carcinogenicity costs $1 million to $3 million and requires at least 3 years to complete," reads the summary of a January 1997 meeting of the Scientific Group on Methodologies for the Safety Evaluation of Chemicals. The main topic of the meeting was the development of alternatives to animal research, and the report continues, "More efficient testing methods may reduce the time required to bring new products to the marketplace and increase the amount of useful information that can be obtained."

Most researchers recognize that the humane treatment of animals isn’t only compassionate—it’s also good science. Imagine trying to measure the effect of blood-pressure medication on a dog that hasn’t been walked in days. We now know that animals’ feelings, behavior, and emotions have a profound effect on their physiological functioning—as is the case with humans. Consequently, after strong initial opposition to the Animal Welfare Act, most researchers have come to support it.

The Humane Society of the United States represents one example of how animal protectionists can set reasonably limited goals that promote animal welfare in ways that better serve both animals and humans. "We’ve contacted animal care and use committees and asked them to work with us to identify techniques that cause pain and distress and figure out ways to share ways to eliminate that in research," says HSUS’s Andrew Rowan. "Some of the committees are rather suspicious; they see a hidden attempt to stop all animal research. The response has been slight so far. But we think that most researchers are bright people and will understand that our primary goal is just to eliminate animal suffering wherever possible."

Norman Letvin, who frequently debates animal protectionists, knows that there are many who would like to end the practice of animal research for good. Although he is ready and willing to discuss the morality and ethics of his work, he thinks that calling an end to the practice would hurt society enormously.

"It is very easy to take an absolutist position and say it is wrong to cause the death of another living animal," Letvin says. "The difficulty in what [researchers] do comes in saying, ‘I understand that what I’m doing is causing the death of a limited number of animals, but I’m making a judgment that the information gained from this limited, focused experiment will yield results that will justify doing the study.’ Many humans infected with viruses or suffering from cancer or heart disease enter into studies that allow the development of new therapeutics. Every day, thousands of humans say, ‘It is worth it for me to be involved in those studies because, even though I probably won’t benefit, others will.’ In the end, the decisions I’m making with respect to experimental animals are not dissimilar."

As we walked to a new facility on Longwood Avenue, Arthur Lage reminded me that it was the former site of Angell Memorial Animal Hospital, which has since moved to Huntington Avenue in Jamaica Plain. He points out where horses were tethered in the courtyard as they waited to be seen by a veterinarian. He indicates a barely visible tower protruding from the rear roof where distempered dogs were once quarantined. "It was hard work," he recalls, somewhat wistfully, of the internship and residency he served at Angell. "But it was rewarding. You might sit up all night with a sick dog or cat, trying to save its life."

Today, Lage cannot devote as much time to saving animals’ lives. Instead, as he says, he’s helping save human lives through animal research, while ensuring that animals are used humanely. Embodied in his work are many of the contradictions that many of us feel when we consider the millions of animals—from mice to monkeys—that annually give their lives for human health. The use of animals in research will not end today, nor tomorrow, but opinions on the matter appear to be evolving, perhaps toward a better life for animals in the laboratory, and toward better science.

John Lauerman used to write the magazine’s " Harvard Health " column. He is coauthor of a book on diabetes and, with Thomas Perls, M.P.H. ’93, M.D., and Margery H. Silver, Ed.D. ’82, of  Living to 100,  forthcoming from Basic Books in March.

You might also like

Psychedelics for Healing

Harvard’s I. Glenn Cohen on the use of “magic mushrooms” and other psychedelics in medicine

The Cost of Political Violence

A Harvard discussion on increasing threats and how to stop them

Former Women’s Hockey Coach Sues Harvard

Katey Stone alleges gender bias in handling of abuse allegations that led to her retirement.

Most popular

Harvard’s Ties to Slavery

A long-awaited, unsparing picture 

Death Throes

Sibling scholars Carol Steiker and Jordan Steiker seek to change how America thinks about capital punishment.

Mechanical Intelligence and Counterfeit Humanity

Reflections on six decades of relations with computers

More to explore

Broadway Director from Harvard Adapting Disney

Broadway music director Madeline Benson on art and collaboration

University People

How Political Tension on Campus Creates Risk Aversion

How overheated political attention warps campus life

The College Pump

Harvard Professor on Social Psychology for Understanding War

Two scholars’ extracurricular efforts in the Middle East

Ethics of Medical Research with Animals

U.S. Law and Animal Experimentation: A Critical Primer

Every country’s law permits medical experimentation on animals. While some countries protect particular kinds of animals from being subject to experimentation—notably great apes and endangered species—very few place concrete limitations on what researchers may cause animals to suffer, given sufficient scientific justification. What laws do, instead, is establish standards for the humane treatment and housing of animals in labs, and they encourage researchers to limit or seek alternatives to the use of animals, when doing that is consistent with the scientific goals of their research. The result, of course, is that no existing regulatory scheme is satisfactory to opponents of animal research. The law, in their view, does nothing more than make the animal research scientist into a sort of James Bond villain: superficially polite, offering fine housing and well-prepared cuisine even to those whom he intends, eventually, to kill.

Of course, the goals of animal experimentation law seem much more reasonable if one accepts that research on animals is both important for medical progress and morally permissible. On those assumptions, it makes a great deal of sense for the law to aim primarily at limiting unnecessary animal suffering even as it licenses scientifically justified experimentation. U.S. law accepts those assumptions and adopts that aim.

The system that has evolved in the United States combines elements of sometimes competing regulatory philosophies. The result is a complex, multilayered system that addresses the most important concerns, but, partly because of historical accident, also leaves some gaps. Even proponents of medical research on animals can see obvious ways in which the regulatory structure could be changed to benefit animals. Perhaps more important, though, is the fact that the existing regulatory structure, imperfect though it may be, is elastic enough to accommodate substantial changes that could reduce unnecessary animal suffering.

Multiple Regulatory Approaches

Animal welfare laws must address three main ways in which unnecessary animal suffering can occur in the context of medical experimentation. First, such suffering can occur when a given research protocol is not well justified scientifically. An experiment that was so badly designed that it could never generate any useful scientific knowledge would never warrant animal suffering. Harder cases result when the amount of suffering is ratcheted down, or the experiment’s potential to generate useful knowledge is ratcheted up. A legal regime concerned with avoiding this kind of unnecessary suffering can opt to trust in the judgment of each individual research scientist, or empower someone besides the researcher to make at least some baseline assessment of the scientific value of each new animal research protocol. It can also provide information and guidance to researchers or overseers to improve their decisions.

Second, unnecessary suffering can occur when the amount of animal suffering induced by an experiment is not strictly required to conduct the experiment—perhaps because more animals are used than are necessary; or because less sentient animals could be substituted for more sentient ones, or computer or tissue models substituted for animals entirely; or because crude experimental procedures are producing avoidable stress or pain. A legal framework seeking to avoid these kinds of unnecessary suffering will encourage or require researchers to use the three Rs: reduce (the number of animals used in experiments), replace (animals with nonanimals, higher-order animals with lower), and refine (experimental procedures causing pain or distress). [1]

Third, unnecessary suffering can occur outside the actual research protocol yet still in the research setting because of inappropriate animal handling, housing, and feeding practices. A legal regime seeking to avoid this kind of suffering will dictate humane standards for animal housing and care.

Given these goals, what sort of regulatory scheme would be best at realizing them? One can imagine a variety of available approaches, from strong, centralized state regulation and monitoring of all experimentation to a hands-off reliance on professional self-regulation among laboratory researchers. On the world stage, the United Kingdom is closest to taking the former approach, Japan to the latter. U.S. law falls somewhere in the middle, in part because U.S. law in this area is in fact the result of a gradual, decades-long merging of the government regulatory and professional self-regulatory approaches. [2]

The government regulatory approach is embodied in the sprawling, strange, and often amended Animal Welfare Act of 1966. In its original form, the AWA was designed to control pet breeding and sale practices; it was passed, in part, as a result of public outcry about the mistreatment of dogs sold to laboratories. As amended, it governs the treatment of animals in a wide range of settings, from pet shops to circuses and from zoos to laboratories. Its enforcement is delegated to the U.S. Department of Agriculture’s Animal and Plant Health Inspection Service, whose inspectors make unannounced site visits to research facilities. Violations uncovered on such visits can result in fines and even, in extreme cases, criminal prosecution. The most common complaint about enforcement under the AWA is that it is rigid and mechanistic.

Because of its historical roots in concern for pets, the AWA’s reach is confined to warm-blooded animals, and it contains special regulations addressed to certain animal favorites: dogs, cats, rabbits, and monkeys. Its animal experimentation regulations apply to any school or research facility that purchases or transports live animals in interstate commerce or that receives federal funding. But in fact the law has never reached the bulk of warm-blooded animals actually used in research. Concern about high regulatory costs—and about possible delay in creating guidelines for other, more popular animals—led the USDA to exclude laboratory rats and mice from its oversight from as early as 1970. In spite of lobbying efforts in the 1980s by proanimal groups, a congressional amendment to the AWA in 2002 legally formalized the agency’s longtime practice, excluding rats, mice, and birds from the definition of “animal.” [3]

In general, the law and its implementing regulations have focused on setting demanding, detailed standards for animal housing and basic standards for pain control. It supports only minimal review of the scientific merit of research protocols, but it requires researchers to make efforts to “reduce, replace, and refine.”

The self-regulatory approach to animal research regulation is embodied in the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals . [4]   The Guide has existed in some version since 1963, when it was introduced as a voluntary set of professional standards for laboratory animal research. Today, the Guide’s standards are mandatory for all research facilities receiving federal funds. The Guide covers the treatment of all vertebrates, which means that, at least in federally funded research, it closes many of the gaps left open by the AWA. Not only are rats, mice, and birds covered, but also cold-blooded vertebrates like zebra fish—currently the go-to animal for laboratory studies of pain and nerve function.

The change in the Guide’s status to a rulebook has altered its content somewhat. Earlier editions’ expansive aspirational goals have given way in later editions to more readily applicable rules. There has also been considerable pressure to get the AWA’s regulatory requirements and the Guide’s standards to match, since all federally funded researchers are bound by both. Indeed, today, the two sets of standards are, if not identical, at least compatible with one another. But in general, where the AWA regulations are more rigidly prescriptive, the Guide permits lab veterinarians to use their professional judgment in applying general standards to particular species or protocols.

Clearly there is room for reform. If the AWA were amended to include rats, mice, and birds, for example, that would be a major step toward ensuring the humane treatment of all animals in public and private labs.  

Federal standards are full of specific requirements for different kinds of studies, but in general, it is fair to say that they offer the most concrete guidance on questions of animal housing and care. The regulations include detailed discussions of square footage, exercise requirements, room temperature, and more. Considerably less guidance is offered on issues of protocol evaluation and implementation of the three Rs.

Of course, this is exactly what might be expected given the incredible volume and variety of animal research in the United States. A central authority can say a lot about how to house and feed monkeys, mice, and zebra fish, and expert advice on those issues will apply to all monkeys, mice, and zebra fish in every lab, no matter what protocols they are being used for. But questions about the other possible sources of unnecessary animal suffering—the scientific justification of a given protocol, or the ways in which animal suffering connected to a given protocol might be avoided or reduced—are too numerous and varied to be answerable in advance by a central authority. With regard to those highly fact-specific questions, U.S. law relies on the expert judgment of local IACUCs.

It is no coincidence that this kind of reliance on decentralized expert committees is also the salient feature of U.S. law governing research on human subjects. The federal Common Rule, [5]  faced with a similar diversity of research protocols to evaluate, regulate, and modify, uses the same tactics as the AWA: it mandates creating research oversight committees (institutional review boards), specifies that their membership should include both relevant expertise and community representation, and empowers them to make and enforce a range of judgments about particular experimental protocols.

While the many IACUCs are expected to exercise independent judgment with regard to the scientific issues brought before them, the U.S. government does its best to inform the judgment by providing them with educational resources. The Public Health Service and the Department of Agriculture Web sites are full of guidance documents and educational resources for laboratory researchers and for IACUC members. There are documents, for example, with specific ideas about how and when to substitute lower-order animals for higher-order animals, and other documents providing up-to-date scientific news about newly developed computer models that can substitute, in some cases, for animal experimentation.

Finally, just as in the human subjects research world, federal regulations are quite commonly supplemented by private education and accreditation. Many research facilities seek accreditation by the Association for the Assessment and Accreditation of Laboratory Animal Care, a professional association of veterinarians and laboratory scientists. AAALAC provides education and does prearranged site inspections of labs once every three years. Educational and inspection standards are built largely around the requirements of the Guide , and the NIH accepts AAALAC accreditation as prima facie evidence of a facility’s compliance with the Guide’s requirements.

Toward Reform: Accountability, Uniformity, Balance

The system of decentralized oversight by local IACUCs has several obvious advantages: it permits oversight by people with knowledge of the local researchers and laboratory facilities; it allows IACUCs to develop specialized knowledge, well tailored to the research being done at their facilities; and it is likely more speedy than any alternative program of centralized governmental research oversight would be. On the other hand, the decentralization of oversight has given rise to a number of problems—which, not surprisingly, are similar to those that beset the IRB system in human subjects research.

First, there is a problem of transparency and accountability. IACUCs are for the most part fairly anonymous. Hardly anyone not directly involved in animal research knows that they exist, much less who their members are. And of course, their members are not elected or in any other way publicly accountable for the decisions they make. Most IACUC decisions do not take the form of opinions or any other form of substantive, publishable decision, but of recommendations to researchers for piecemeal alteration of protocols. A central repository of IACUC minutes, and of policies adopted by different IACUCs, might both increase accountability and stimulate new ideas by creating cross talk between IACUCs. But any such repository would have to be created with an eye toward preserving researchers’ intellectual property.

Second, decentralization almost necessarily gives rise to a lack of uniformity in decision-making and in quality of research oversight. One IACUC may conclude that a protocol involves unnecessarily harsh treatment of animals or presents an opportunity for substitution of nonanimal models; another may view the original protocol as unproblematic and requiring no amendment. A number of studies have shown that similar protocols are treated quite differently by different IACUCs. [6]  It is unclear what the implications of such findings are. Do they reveal that IACUCs have differing standards relating to animal welfare? That they judge similar protocols differently when they are presented by different researchers? Or some combination of these factors? In any case, enforced uniformity across IACUCs is a dangerous solution to propose for the problem of varying standards, in the absence of clear knowledge about whose standards are appropriate—and whose would be enforced.

A third complaint about the decentralized approach to animal-research regulation involves the perception that the U.S. government is too deferential to local IACUCs and does not take the task of auditing labs sufficiently seriously. In the early 2000s, there were some high-profile allegations made by whistleblowers from the USDA’s Animal and Plant Health Inspection Service (APHIS) that audit findings were deliberately being watered down to be less critical than the field officers originally intended them to be. [7]  U.S. audits of APHIS confirmed allegations of lax auditing in some regions of the country. [8]  The obvious reform here is to better fund and train both the regulatory overseers and those who audit their performance.

There are other important criticisms of the U.S. regulatory regime not directly connected to its choice of decentralized decision-making. First, there is the question of scientific justification for animal suffering. The AWA does not ask IACUCs to balance animal suffering against the scientific merit or promise of any given experiment. Instead, it asks IACUCs to ensure only that any given protocol has scientific merit and that any animal suffering the protocol induces is strictly necessary to that science. The result is that any study that will advance science, even in a very small way, can be used to justify tremendous amounts of animal suffering, as long as the suffering is necessary to the advance. Though they do seek to modify studies via use of the three Rs, IACUCs almost never reject protocols.

Finally, and most importantly, there is the issue of which animals are protected. As already mentioned, the hundreds of thousands of rats, mice, and birds used in private, nonfederally funded labs are not subject to any federal regulation. (Some individual states’ anticruelty statutes may apply in some cases, but there is very limited case law in the area.) Excluded, also, are cold-blooded animals. This means that there is no federal legal pressure on private firms such as drug companies to reduce or refine animal use, or to replace animals with computer or tissue models—a strategy that may be particularly feasible in studies of toxicology or drug metabolization.

Even in federally funded facilities, the living conditions of rats, mice, and birds are not subject to the USDA’s APHIS inspection; only in AAALAC-accredited facilities is there oversight beyond self-reporting, and AAALAC does scheduled inspections only once every three years. Rats and mice, it should be stressed, are the most commonly used laboratory animals. In addition, U.S. law offers no protection for invertebrate, cold-blooded animals such as cephalopods. By contrast, Europe has recently moved to protect cephalopods in light of their manifest intelligence and sentience. Nor does U.S. law prevent research on great apes, or ban (though it does regulate) the use of wild-caught animals. And the United States is one of only two governments in the world that still permits invasive research on chimpanzees, though the scope of federal funding for chimp research has recently been sharply limited. [9]  (See “Raising the Bar: The Implications of the IOM Report on the Use of Chimpanzees in Research,” in this volume.)

Clearly there is room for reform. Some needed reform involves stepping up research oversight. If the AWA were amended to include rats, mice, and birds, for example, that would be a major step toward ensuring the humane treatment of all animals in public and private labs. In addition, the inspection rate for facilities could be more frequent. Publicly funded U.S. labs are inspected by APHIS about once a year, by their own IACUCs twice a year, and by AAALAC (if they choose to be AAALAC-certified) once every three years. Compare this to the U.K. system of inspecting about once a month. Other reforms could involve improving rigid and not-terribly-useful existing regulations, like cage-size requirements currently based on animals’ body size rather than on their behavioral needs. Most significantly, the law could be reformed to permit a more explicit balancing of harms to animals (including both suffering and death) against the scientific gains at which the research aims. Empowering IACUCs to engage in such balancing is hardly radical; IRBs, for example, are already empowered to engage in such balancing in the human subjects research area, and this has not caused research to grind to a halt. Such a reform would require us to confront directly the question of how much suffering humans can impose on other species in return for small but real gains in knowledge.

Finally, a great deal can be accomplished even within an unchanged legal regime. The most urgent need is for more to be done to implement the three Rs. The familiar calls for better education about replacement techniques and more aggressive IACUC intervention on behalf of reduction and refinement are, of course, well justified. But even more dramatic reduction might be achieved if the goal of reduction were pursued not only within but also across protocols. There might be significant gains from putting animal-sharing procedures in place at the institutional level. At the moment, animals are commonly euthanized whenever the particular research project they’re involved in comes to an end, without regard to the animal’s age or health status. If a protocol involves attempts to breed, for example, mice with particular genetic traits, the pups born without those traits are routinely euthanized. If research facilities could work with researchers to use healthy animals from one study in another, rather than default to their euthanization, then fewer animals would need to be bred for suffering.

Stephen R. Latham is director of the Interdisciplinary Center for Bioethics at Yale University. He has published on a broad range of issues at the intersection of bioethics and law. He is a former board member of the American Society for Bioethics and Humanities, a former graduate fellow of Harvard’s Safra Center on Ethics, and a former research fellow of the University of Edinburgh’s Institute for Advanced Studies in Humanities. His current research includes a project funded by the Robert Wood Johnson Foundation to create a database of state statutes and cases criminalizing HIV exposure and a project on a legal framework for newborn whole-exome screening. 

• 1. The widely accepted “Three Rs” terminology was first introduced into the animal research literature in W.M.S. Russell and R.L. Burch, The Principals of Human Experimentation Technique (London: Methuen, 1959). ↵
• 2. A detailed account of the confluence of these two streams of regulation (to which my brief discussion here is heavily indebted) is provided by L. Carbone, What Animals Want: Expertise and Advocacy in Laboratory Animal Welfare Policy (Oxford, U.K.: Oxford University Press, 2004), p. 34ff. ↵
• 3. Wild-caught rats and mice are included in the regulations. For more detail, see Carbone, What Animals Want , p. 69ff. ↵
• 4. National Research Council, Guide for the Care and Use of Laboratory Animals , 8th ed.(Washington, D.C.: National Academies Press, 2011). ↵
• 5. U.S. Department of Health and Human Services, 45 CFR 46. ↵
• 6. See, for example, S. Plous and H. Herzog, “Reliability of Protocol Reviews for Animals Research,” Science 293 (2001): 608-9. ↵
• 7. See, for example, the statement of Dr. Isis Johnson-Brown, USDA whistleblower, alleging regulatory inaction on her report criticizing cage conditions at the Oregon Primate Center, at http://www.all-creatures.org/saen/articles-statementofijb.html, accessed October 2, 2012. ↵
• 8. USDA Office of Inspector General, Western Region, “Audit Report: APHIS Animal Care Program Inspection and Enforcement Activities,” Report No. 33002-3-SF, September 2005, p. i, http://www.usda.gov/oig/webdocs/33002-03-SF.pdf. ↵
• 9. See Institute of Medicine, Committee on the Use of Chimpanzees in Biomedical and Behavioral Research: Assessing the Necessity (Washington, D.C.: National Academies Press, 2011); B.M. Altevogt et al., “Guiding Limited Use of Chimpanzees in Research,” Science 335 (2012): 41-42. ↵

Loading metrics

Open Access

Essays articulate a specific perspective on a topic of broad interest to scientists.

See all article types »

A guide to open science practices for animal research

Contributed equally to this work with: Kai Diederich, Kathrin Schmitt

Affiliation German Federal Institute for Risk Assessment, German Centre for the Protection of Laboratory Animals (Bf3R), Berlin, Germany

* E-mail: [email protected]

• Kai Diederich, 
• Kathrin Schmitt, 
• Philipp Schwedhelm, 
• Bettina Bert, 
• Céline Heinl

Published: September 15, 2022

• https://doi.org/10.1371/journal.pbio.3001810
• Reader Comments

Translational biomedical research relies on animal experiments and provides the underlying proof of practice for clinical trials, which places an increased duty of care on translational researchers to derive the maximum possible output from every experiment performed. The implementation of open science practices has the potential to initiate a change in research culture that could improve the transparency and quality of translational research in general, as well as increasing the audience and scientific reach of published research. However, open science has become a buzzword in the scientific community that can often miss mark when it comes to practical implementation. In this Essay, we provide a guide to open science practices that can be applied throughout the research process, from study design, through data collection and analysis, to publication and dissemination, to help scientists improve the transparency and quality of their work. As open science practices continue to evolve, we also provide an online toolbox of resources that we will update continually.

Citation: Diederich K, Schmitt K, Schwedhelm P, Bert B, Heinl C (2022) A guide to open science practices for animal research. PLoS Biol 20(9): e3001810. https://doi.org/10.1371/journal.pbio.3001810

Copyright: © 2022 Diederich et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The authors received no specific funding for this work.

Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: All authors are employed at the German Federal Institute for Risk Assessment and part of the German Centre for the Protection of Laboratory Animals (Bf3R) which developed and hosts animalstudyregistry.org , a preregistration platform for animal studies and animaltestinfo.de, a database for non-technical project summaries (NTS) of approved animal study protocols within Germany.

Abbreviations: CC, Creative Commons; CIRS-LAS, critical incident reporting system in laboratory animal science; COVID-19, Coronavirus Disease 2019; DOAJ, Directory of Open Access Journals; DOI, digital object identifier; EDA, Experimental Design Assistant; ELN, electronic laboratory notebook; EU, European Union; IMSR, International Mouse Strain Resource; JISC, Joint Information Systems Committee; LIMS, laboratory information management system; MGI, Mouse Genome Informatics; NC3Rs, National Centre for the Replacement, Refinement and Reduction of Animals in Research; NTS, non-technical summary; RRID, Research Resource Identifier

Introduction

Over the past decade, the quality of published scientific literature has been repeatedly called into question by the failure of large replication studies or meta-analyses to demonstrate sufficient translation from experimental research into clinical successes [ 1 – 5 ]. At the same time, the open science movement has gained more and more advocates across various research areas. By sharing all of the information collected during the research process with colleagues and with the public, scientists can improve collaborations within their field and increase the reproducibility and trustworthiness of their work [ 6 ]. Thus, the International Reproducibility Networks have called for more open research [ 7 ].

However, open science practices have not been adopted to the same degree in all research areas. In psychology, which was strongly affected by the so-called reproducibility crisis, the open science movement initiated real practical changes leading to a broad implementation of practices such as preregistration or sharing of data and material [ 8 – 10 ]. By contrast, biomedical research is still lagging behind. Open science might be of high value for research in general, but in translational biomedical research, it is an ethical obligation. It is the responsibility of the scientist to transparently share all data collected to ensure that clinical research can adequately evaluate the risks and benefits of a potential treatment. When Russell and Burch published “The Principles of Humane Experimental Technique” in 1959, scientists started to implement their 3Rs principle to answer the ethical dilemma of animal welfare in the face of scientific progress [ 11 ]. By replacing animal experiments wherever possible, reducing the number of animals to a strict minimum, and refining the procedures where animals have still to be used, this ethical dilemma was addressed. However, in recent years, whether the 3Rs principle is sufficient to fully address ethical concerns about animal experiments has been questioned [ 12 ].

Most people tolerate the use of animals for scientific purposes only under the basic assumption that the knowledge gained will advance research in crucial areas. This implies that performed experiments are reported in a way that enables peers to benefit from the collected data. However, recent studies suggest that a large proportion of animal experiments are never actually published. For example, scientists working within the European Union (EU) have to write an animal study protocol for approval by the competent authorities of the respective country before performing an animal experiment [ 13 ]. In these protocols, scientists have to describe the planned study and justify every animal required for the project. By searching for publications resulting from approved animal study protocols from 2 German University Medical Centers, Wieschowski and colleagues found that only 53% of approved protocols led to a publication after 6 years [ 14 ]. Using a similar approach, Van der Naald and colleagues determined a publication rate of 60% at the Utrecht Medical Center [ 15 ]. In a follow-up survey, the respective researchers named so-called “negative” or null-hypothesis results as the main cause for not publishing outcomes [ 15 ]. The current scientific system is shaped by publishers, funders, and institutions and motivates scientists to publish novel, surprising, and positive results, revealing one of the many structural problems that the numerous efforts towards open science initiatives are targeting. Non-publication not only strongly contradicts ethical values, but also it compromises the quality of published literature by leading to overestimation of effect sizes [ 16 , 17 ]. Furthermore, publications of animal studies too often show poor reporting that strongly impairs the reproducibility, validity, and usefulness of the results [ 18 ]. Unfortunately, the idea that negative or equivocal findings can also contribute to the gain of scientific knowledge is frequently neglected.

So far, the scientific community using animals has shown limited resonance to the open science movement. Due to the strong controversy surrounding animal experiments, scientists have been reluctant to share information on the topic. Additionally, translational research is highly competitive and researchers tend to be secretive about their ideas until they are ready for publication or patent [ 19 , 20 ]. However, this missing openness could also point to a lack of knowledge and training on the many open science options that are available and suitable for animal research. Researchers have to be convinced of the benefits of open science practices, not only for science in general, but also for the individual researcher and each single animal. Yet, the key players in the research system are already starting to value open science practices. An increasing number of journals request open sharing of data, funders pay for open access publications and institutions consider open science practices in hiring decisions. Open science practices can improve the quality of work by enabling valuable scientific input from peers at the early stages of research projects. Furthermore, the extended communication that open science practices offer can draw attention to research and help to expand networks of collaborators and lead to new project opportunities or follow-up positions. Thus, open science practices can be a driver for careers in academia, particularly those of early career researchers.

Beyond these personal benefits, improving transparency in translational biomedical research can boost scientific progress in general. By bringing to light all the recorded research outputs that until now have remained hidden, the publication bias and the overestimation of effect sizes can be reduced [ 17 ]. Large-scale sharing of data can help to synthesize research outputs in preclinical research that will enable better decision-making for clinical research. Disclosing the whole research process will help to uncover systematic problems and support scientists in thoroughly planning their studies. In the long run, we predict that the implementation of open science practices will lead to the use of fewer animals in unintentionally repeated experiments that previously showed unreported negative results or in the establishment of methods by avoiding experimental dead ends that are often not published. More collaborations and sharing of materials and methods can further reduce the number of animal experiments used for the implementation of new techniques.

Open science can and should be implemented at each step of the research process ( Fig 1 ). A vast number of tools are already provided that were either directly conceptualized for animal research or can be adapted easily. In this Essay, we provide an overview of open science tools that improve transparency, reliability, and animal welfare in translational in vivo biomedical research by supporting scientists to clearly communicate their research and by supporting collaborative working. Table 1 lists the most prominent open science tools we discuss, together with their respective links. We have structured this Essay to guide you through which tools can be used at each stage of the research process, from planning and conducting experiments, through to analyzing data and communicating the results. However, many of these tools can be used at many different steps. Table 1 has been deposited on Zenodo and will be updated continuously [ 21 ].

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

Application of open science practices at each step of the research process can maximize the impact of performed animal experiments. The implementation of these practices will lead to less time pressure at the end of a project. Due to the connection of most of these open science practices, spending more time in the planning phase and during the conduction of experiments will save time during the data analysis and publication of the study. Indeed, consulting reporting guidelines early on, preregistering a statistical plan, and writing down crucial experimental details in an electronic lab notebook, will strongly accelerate the writing of a manuscript. If protocols or even electronic lab notebooks were made public, just citing these would simplify the writing of publications. Similarly, if a data management plan is well designed before starting data collection, analyzing, and depositing data in a public repository, as is increasingly required, will be fast. NTS, non-technical summary.

https://doi.org/10.1371/journal.pbio.3001810.g001

https://doi.org/10.1371/journal.pbio.3001810.t001

Planning the study

Transparent practices can be adopted at every stage of the research process. However, to ensure full effectivity, it is highly recommended to engage in detailed planning before the start of the experiment. This can prevent valuable time from being lost at the end of the study due to careless decisions being made at the beginning. Clarifying data management at the start of a project can help avoiding filing chaos that can be very time consuming to untangle. Keeping clear track of a project and study design will also help if new colleagues are included later on in the project or if entire project parts are handed over. In addition, all texts written on the rationale and hypothesis of the study or method descriptions, or design schemes created during the planning phase can be used in the final publications ( Fig 1 ). Similarly, information required for preregistration of animal studies or for reporting according to the ARRIVE guidelines are an extension of the details required for ethical approval [ 22 , 23 ]. Thus, the time burden within the planning phase is often overestimated. Furthermore, the thorough planning of experiments can avoid the unnecessary use of animals by preventing wrong avenues from being pursued.

Implementing open scientific practices at the beginning of a project does not mean that the idea and study plan must be shared immediately, but rather is critical for making the entire workflow transparent at the end of the project. However, optional early sharing of information can enable peers to give feedback on the study plan. Studies potentially benefit more from this a priori input than they would from the classical a posteriori peer-review process.

Most people perceive guidelines as advice that instructs on how to do something. However, it is sometimes useful to consider the term in its original meaning; “the line that guides us”. In this sense, following guidelines is not simply fulfilling a duty, but is a process that can help to design a sound research study and, as such, guidelines should be consulted at the planning stage of a project. The PREPARE guidelines are a list of important points that should be thought-out before starting a study involving animal experiments in order to reduce the waste of animals, promote alternatives, and increase the reproducibility of research and testing [ 24 ]. The PREPARE checklist helps to thoroughly plan a study and focuses on improving the communication and collaboration between all involved participants of the study (i.e., animal caretakers and scientists). Indeed, open science begins with the communication within a research facility. It is currently available in 33 languages and the responsible team from Norecopa, Norway’s 3R-center, takes requests for translations into further languages.

The UK Reproducibility Network has also published several guiding documents (primers) on important topics for open and reproducible science. These address issues such as data sharing [ 25 ], open access [ 26 ], open code and software [ 27 ], and preprints [ 28 ], as well as preregistration and registered reports [ 27 ]. Consultation of these primers is not only helpful in the relevant phases of the experiment but is also encouraged in the planning phase.

Although the ARRIVE guidelines are primarily a reporting guideline specifically designed for preparing a publication containing animal data, they can also support researchers when planning their experiments [ 22 , 23 ]. Going through the ARRIVE website, researchers will find tools and explanations that can support them in planning their experiments [ 29 ]. Consulting the ARRIVE checklist at the beginning of a project can help in deciding what details need to be documented during conduction of the experiments. This is particularly advisable, given that compliance to ARRIVE is still poor [ 18 ].

Experimental design

To maximize the validity of performed experiments and the knowledge gained, designing the study well is crucial. It is important that the chosen animal species reflects the investigated disease well and that basic characteristics of the animal, such as sex or age, are considered carefully [ 30 ]. The Canadian Institutes of Health Research provides a collection of resources on the integration of sex and gender in biomedical research with animals, including tips and tools for researchers and reviewers [ 31 ]. Additionally, it is advisable to avoid unnecessary standardization of biological and environmental factors that can reduce the external validity of results [ 32 ]. Meticulous statistical planning can further optimize the use of animals. Free to use online tools for calculating sample sizes such as G*Power or the inVivo software package for R can further support animal researchers in designing their statistical plan [ 33 , 34 ]. Randomization for the allocation of groups can be supported with specific tools for scientists like Research Randomizer, but also by simple online random number generators [ 35 ]. Furthermore, it might be advisable when designing the study to incorporate pathological analyses into the experimental plan. Optimal planning of tissue collection, performance of pathological procedures according to accepted best practices, and use of optimal pathological analysis and reporting methods can add some extra knowledge that would otherwise be lost. This can improve the reproducibility and quality of translational biomedicine, especially, but not exclusively, in animal studies with morphological endpoints. In all animal studies, unexpected deaths in experimental animals can occur and be the cause of lost data or missed opportunities to identify health problems [ 36 , 37 ].

To support researchers in designing their animal research, the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs) has also developed the Experimental Design Assistant (EDA) [ 38 , 39 ]. This online tool helps researchers to better structure in vivo research by creating detailed schemes of the study design. It provides feedback on the entered design, drawing researcher’s attention to crucial decisions in the project. The resulting schemes can be used to transparently share the study design by uploading it into a study preregistration, enclosing it in a grant application, or submitting it with a final manuscript. The EDA can be used for different study designs in diverse scenarios and helps to communicate researcher plans to others [ 40 ]. The EDA might be particularly of interest to clarify very complex study designs involving multiple experimental groups. Working with the EDA might appear rather complex in the beginning, but the NC3R provides regular webinars that can help to answer any questions that arise.

Preregistration

Preregistration is an effective tool to improve the quality and transparency of research. To preregister their work, scientists must determine crucial details of the study before starting any experiment. Changes occurring during a study can be outlined at the end. A preregistered study plan should include at least the hypothesis and determine all the parameters that are known in advance. A description of the planned study design and statistical analysis will enable reviewers and peers to better retrace the workflow. It can prevent the intentional use of the flexibility of analysis to reach p -values under a certain significance level (e.g., p-hacking or HARKing (Hypothesizing After Results are Known)). With preregistration, scientists can also claim their idea at an early stage of their research with a citable individual identifier that labels the idea as their own. Some open preregistration platforms also provide a digital object identifier (DOI), which makes the registered study citable. Three public registries actively encourage the preregistration of animal studies conducted around the world: OSF registry, preclinicaltrials.eu, and animalstudyregistry.org [ 41 – 45 ]. Scientists can choose the registry according to their needs. Preregistering a study in a public registry supports scientists in planning their study and later to critically reevaluate their own work and assess its limitations and potentials.

As an alternative to public registries, researchers can also submit their study plan to one of hundreds of journals already publishing registered reports, including many journals open to animal research [ 8 ]. A submitted registered report passes 2 steps of peer review. In the first step, reviewers comment on the idea and the study design. After an “in-principle-acceptance,” researchers can conduct their study as planned. If the authors conduct the experiments as described in the accepted study protocol, the journal will publish the final study regardless of the outcome. This might be an attractive option, especially for early career researchers, as a manuscript is published at the beginning of a project with the guarantee of a future final publication.

The benefits of preregistration can already be observed in clinical research, where registration has been mandatory for most trials for more than 20 years. Preregistration in clinical research has helped to make known what has been tested and not just what worked and was published, and the implementation of trial registration has strongly reduced the number of publications reporting significant treatment effects [ 46 ]. In animal research, with its unrealistically high percentage of positive results, preregistration seems to be particularly worthwhile.

Research data management

To get the most out of performed animal experiments, effective sharing of data at the end of the study is essential. Sharing research data optimally is complex and needs to be prepared in advance. Thus, data management can be seen as one part of planning a study thoroughly. Many funders have recognized the value of the original research data and request a data management plan from applicants in advance [ 25 , 47 ]. Various freely available tools such as DMPTool or DMPonline already exist to design a research data management plan that complies to the requirements of different funders [ 48 , 49 ]. The data management plan defines the types of data collected and describes the handling and names responsible persons throughout the data lifecycle. This includes collecting the data, analyzing, archiving, and sharing it. Finally, a data management plan enables long-term access and the possibility for reuse by peers. Developing such a plan, whether it is required by funders or not, will later simplify the application of the FAIR data principle (see section on the FAIR data principle). The Longwood Medical Area Research Data Management Working Group from the Harvard Medical School developed a checklist to assist researchers in optimally managing their data throughout the data lifecycle [ 50 ]. Similarly, the Joint Information Systems Committee (JISC) provides a great research data management toolkit including a checklist for researchers planning their project [ 51 ]. Consulting this checklist in the planning phase of a project can prevent common errors in research data management.

Non-technical project summary

One instrument specifically conceived to create transparency on animal research for the general public is the so-called non-technical project summary (NTS). All animal protocols approved within the EU must be accompanied by these comprehensible summaries. NTSs are intended to inform the public about ongoing animal experiments. They are anonymous and include information on the objectives and potential benefits of the project, the expected harm, the number of animals, the species, and a statement of compliance with the requirements of the 3Rs principle. However, beyond simply informing the public, NTSs can also be used for meta-research to help identify new research areas with an increased need for new 3R technologies [ 52 , 53 ]. NTSs become an excellent tool to appropriately communicate the scientific value of the approved protocol and for meta-scientists to generate added value by systematically analyzing theses summaries if they fulfill a minimum quality threshold [ 54 , 55 ]. In 2021, the EU launched the ALURES platform ( Table 1 ), where NTSs from all member states are published together, opening the opportunities for EU-wide meta-research. NTSs are, in contrast to other open science practices, mandatory in the EU. However, instead of thinking of them as an annoying duty, it might be worth thoroughly drafting the NTS to support the goals of more transparency towards the public, enabling an open dialogue and reducing extreme opinions.

Conducting the experiments

Once the experiments begin, documentation of all necessary details is essential to ensure the transparency of the workflow. This includes methodological details that are crucial for replicating experiments, but also failed attempts that could help peers to avoid experiments that do not work in the future. All information should be stored in such a way that it can be found easily and shared later. In this area, many new tools have emerged in recent years ( Table 1 ). These tools will not only make research transparent for colleagues, but also help to keep track of one’s own research and improve internal collaboration.

Electronic laboratory notebooks

Electronic laboratory notebooks (ELNs) are an important pillar of research data management and open science. ELNs facilitate the structured and harmonized documentation of the data generation workflow, ensure data integrity, and keep track of all modifications made to the original data based on an audit trail option. Moreover, ELNs simplify the sharing of data and support collaborations within and outside the research group. Methodological details and research data become searchable and traceable. There is an extensive amount of literature providing advice on the selection and the implementation process of an ELN depending on the specific needs and research area and its discussion would be beyond the scope of this Essay [ 56 – 58 ]. Some ELNs are connected to a laboratory information management system (LIMS) that provides an animal module supporting the tracking of animal details [ 59 ]. But as research involving animals is highly heterogeneous, this might not be the only decision point and we cannot recommend a specific ELN that is suitable for all animal research.

ELNs are already established in the pharmaceutical industry and their use is on the rise among academics as well. However, due to concerns around costs for licenses, data security, and loss of flexibility, many research institutions still fear the expenses that the introduction of such a system would incur [ 56 ]. Nevertheless, an increasing number of academic institutions are implementing ELNs and appreciating the associated benefits [ 60 ]. If your institution already has an ELN, it might be easiest to just use the option available in the research environment. If not, the Harvard Medical School provides an extensive and updated overview of various features of different ELNs that can support scientists in choosing the appropriate one for their research [ 61 ]. There are many commercial ELN products, which may be preferred when the administrative workload should be outsourced to a large extent. However, open-source products such as eLabFTW or open BIS provide a greater opportunity for customization to meet specific needs of individual research institutions [ 62 – 64 ]. A huge number of options are available depending on the resources and the features required. Some scientists might prefer generic note taking tools such as Evernote or just a simple Word document that offers infinite flexibility, but specific ELNs can further support good record keeping practice by providing immutability, automated backups, standardized methods, and protocols to follow. Clearly defining the specific requirements expected might help to choose an adequate system that would improve the quality of the record compared to classical paper laboratory notebooks.

Sharing protocols

Adequate sharing of methods in translational biomedical sciences is key to reproducibility. Several repositories exist that simplify the publication and exchange of protocols. Writing down methods at the end of the project bears the risk that crucial details might be missing [ 65 ]. On protocols.io, scientists can note all methodological details of a procedure, complete them with uploaded documents, and keep them for personal use or share them with collaborators [ 66 ]. Authors can also decide at any point in time to make their protocol public. Protocols published on protocols.io receive a DOI and become citable; they can be commented on by peers and adapted according to the needs of the individual researcher. Protocol.io files from established protocols can also be submitted together with some context and sample datasets to PLOS ONE , where it can be peer-reviewed and potentially published [ 67 , 68 ]. Depending on the affiliation of the researchers to academia or industry and on an internal or public sharing of files, protocols.io can be free of charge or come with costs. Other journals also encourage their authors to deposit their protocols in a freely accessible repository, such as protocol exchange from Nature portfolio [ 69 ]. Another option might be to separately submit a protocol that was validated by its use in an already published research article to an online and peer-reviewed journal specific for research protocols, such as Bio-Protocol. A multitude of journals, including eLife and Science already collaborate with Bio-Protocol and recommend authors to publish the method in Bio-Protocol [ 70 ]. Bio-Protocol has no submission fees and is freely available to all readers. Both protocols.io and Bio-Protocol allow the illustration of complex scientific methods by uploading videos to published protocols. In addition, protocols can be deposited in a general research repository such as the Open Science Framework (OSF repository) and referenced in appropriate publications.

Sharing critical incidents

Sharing critical or even adverse events that occur in the context of animal experimentation can help other scientists to avoid committing the same mistakes. The system of sharing critical incidents is already established in clinical practice and helps to improve medical care [ 71 , 72 ]. The online platform critical incident reporting system in laboratory animal science (CIRS-LAS) represents the first preclinical equivalent to these clinical systems [ 73 ]. With this web-based tool, critical incidents in animal research can be reported anonymously without registration. An expert panel helps to analyze the incident to encourage an open dialogue. Critical incident reporting is still very marginal in animal research and performed procedures are very variable. These factors make a systemic analysis and a targeted search of incidence difficult. However, it may be of special interest for methods that are broadly used in animal research such as anesthesia. Indeed, a broad feed of this system with data on errors occurring in standard procedures today could help avoid critical incidences in the future and refine animal experiments.

Sharing animals, organs, and tissue

When we think about open science, sharing results and data are often in focus. However, sharing material is also part of a collaborative and open research culture that could help to greatly reduce the number of experimental animals used. When an animal is killed to obtain specific tissue or organs, the remainder is mostly discarded. This may constitute a wasteful practice, as surplus tissue can be used by other researchers for different analyses. More animals are currently killed as surplus than are used in experiments, demonstrating the potential for sharing these animals [ 74 , 75 ].

Sharing information on generated surplus is therefore not only economical, but also an effective way to reduce the number of animals used for scientific purposes. The open-source software Anishare is a straightforward way for breeders of genetically modified lines to promote their surplus offspring or organs within an institution [ 76 ]. The database AniMatch ( Table 1 ) connects scientists within Europe who are offering tissue or organs with scientists seeking this material. Scientists already sharing animal organs can support this process by describing it in publications and making peers aware of this possibility [ 77 ]. Specialized research communities also allow sharing of animal tissue or animal-derived products worldwide that are typically used in these fields on a collaborative basis via the SEARCH-framework [ 78 , 79 ]. Depositing transgenic mice lines into one of several repositories for mouse strains can help to further minimize efforts in producing new transgenic lines and most importantly reduce the number of surplus animals by supporting the cryoconservation of mouse lines. The International Mouse Strain Resource (IMSR) can be used to help find an adequate repository or to help scientists seeking a specific transgenic line find a match [ 80 ].

Analyzing the data

Animal researchers have to handle increasingly complex data. Imaging, electrophysiological recording, or automated behavioral tracking, for example, produce huge datasets. Data can be shared as raw numerical output but also as images, videos, sounds, or other forms from which raw numerical data can be generated. As the heterogeneity and the complexity of research data increases, infinite possibilities for analysis emerge. Transparently reporting how the data were processed will enable peers to better interpret reported results. To get the most out of performed animal experiments, it is crucial to allow other scientists to replicate the analysis and adapt it to their research questions. It is therefore highly recommended to use formats and tools during the analysis that allow a straightforward exchange of code and data later on.

Transparent coding

The use of non-transparent analysis codes have led to a lack of reproducibility of results [ 81 ]. Sharing code is essential for complex analysis and enables other researchers to reproduce results and perform follow-up studies, and citable code gives credit for the development of new algorithms ( Table 1 ). Jupyter Notebooks are a convenient way to share data science pipelines that may use a variety of coding languages, including like Python, R or Matlab, and also share the results of analyses in the form of tables, diagrams, images, and videos. Notebooks contain source code and can be published or collaboratively shared on platforms like GitHub or GitLab, where version control of source code is implemented. The data-archiving tool Zenodo can be used to archive a repository on GitHub and create a DOI for the archive. Thereby contents become citable. Using free and open-source programming language like R or Python will increase the number of potential researchers that can work with the published code. Best practice for research software is to publish the source code with a license that allows modification and redistribution.

Choice of data visualization

Choosing the right format for the visualization of data can increase its accessibility to a broad scientific audience and enable peers to better judge the validity of the results. Studies based on animal research often work with very small sample sizes. Visualizing these data in histograms may lead to an overestimation of the outcomes. Choosing the right dot plots that makes all recorded points visible and at the same time focusses on the summary instead of the individual points can further improve the intuitive understanding of a result. If the sample size is too low, it might not be meaningful to visualize error bars. A variety of freely available tools already exists that can support scientists in creating the most appropriate graphs for their data [ 82 ]. In particular, when representing microscopy results or heat maps, it should be kept in mind that a large part of the population cannot perceive the classical red and green representation [ 83 ]. Opting for the color-blind safe color maps and checking images with free tools such as color oracle ( Table 1 ) can increase the accessibility of graphs. Multiple journals have already addressed flaws in data visualization and have introduced new policies that will accelerate the uptake of transparent representation of results.

Publication of all study outcomes

Open science practices have received much attention in the past few years when it comes to publication of the results. However, it is important to emphasize that although open science tools have their greatest impact at the end of the project, good study preparation and sharing of the study plan and data early on can greatly increase the transparency at the end.

The FAIR data principle

To maximize the impact and outcome of a study, and to make the best long-term use of data generated through animal experiments, researchers should publish all data collected during their research according to the FAIR data principle. That means the data should be findable, accessible, interoperable, and reusable. The FAIR principle is thus an extension of open access publishing. Data should not only be published without paywalls or other access restrictions, but also in such a manner that they can be reused and further processed by others. For this, legal as well as technical requirements must be met by the data. To achieve this, the GoFAIR initiative has developed a set of principles that should be taken into account as early as at the data collection stage [ 49 , 84 ]. In addition to extensively described and machine-readable metadata, these principles include, for example, the application of globally persistent identifiers, the use of open file formats, and standardized communication protocols to ensure that humans and machines can easily download the data. A well-chosen repository to upload the data is then just the final step to publish FAIR data.

FAIR data can strongly increase the knowledge gained from performed animal experiments. Thus, the same data can be analyzed by different researchers and could be combined to obtain larger sample sizes, as already occurs in the neuroimaging community, which works with comparable datasets [ 85 ]. Furthermore, the sharing of data enables other researchers to analyze published datasets and estimate measurement reliabilities to optimize their own data collection [ 86 , 87 ]. This will help to improve the translation from animal research into clinics and simultaneously reduce the number of animal experiment in future.

Reporting guidelines

In preclinical research, the ARRIVE guidelines are the current state of the art when it comes to reporting data based on animal experiments [ 22 , 23 ]. The ARRIVE guidelines have been endorsed by more than 1,000 journals who ask that scientists comply with them when reporting their outcomes. Since the ARRIVE guidelines have not had the expected impact on the transparency of reporting in animal research publications, a more rigorous update has been developed to facilitate their application in practice (ARRIVE 2.0 [ 23 ]). We believe that the ARRIVE guidelines can be more effective if they are implemented at a very early stage of the project (see section on guidelines). Some more specialized reporting guidelines have also emerged for individual research fields that rely on animal studies, such as endodontology [ 88 ]. The equator network collects all guidelines and makes them easily findable with their search tool on their website ( Table 1 ). MERIDIAN also offers a 1-stop shop for all reporting guidelines involving the use of animals across all research sectors [ 89 ]. It is thus worth checking for new reporting guidelines before preparing a manuscript to maximize the transparency of described experiments.

Identifiers

Persistent identifiers for published work, authors, or resources are key for making public data findable by search engines and are thus a prerequisite for compliance to FAIR data principles. The most common identifier for publications will be a DOI, which makes the work citable. A DOI is a globally unique string assigned by the International DOI Foundation to identify content permanently and provide a persistent link to its location on the Internet. An ORCID ID is used as a personal persistent identifier and is recommendable to unmistakably identify an author ( Table 1 ). This will avoid confusions between authors with the same name or in the case of name changes or changes of affiliation. Research Resource Identifiers (RRID) are unique ID numbers that help to transparently report research resources. RRID also apply to animals to clearly identify the species used. RRID help avoid confusion between different names or changing names of genetic lines and, importantly, make them machine findable. The RRID Portal helps scientists find a specific RRID or create one if necessary ( Table 1 ). In the context of genetically altered animal lines, correct naming is key. The Mouse Genome Informatics (MGI) Database is the authoritative source of official names for mouse genes, alleles, and strains ([ 90 ]).

Preprint publication

Preprints have undergone unprecedented success, particularly during the height of the Coronavirus Disease 2019 (COVID-19) pandemic when the need for rapid dissemination of scientific knowledge was critical. The publication process for scientific manuscripts in peer-reviewed journals usually requires a considerable amount of time, ranging from a few months to several years, mainly due to the lengthy review process and inefficient editorial procedures [ 91 , 92 ]. Preprints typically precede formal publication in scientific journals and, thus, do not go through a peer review process, thus, facilitating the prompt open dissemination of important scientific findings within the scientific community. However, submitted papers are usually screened and checked for plagiarism. Preprints are assigned a DOI so they can be cited. Once a preprint is published in a journal, its status is automatically updated on the preprint server. The preprint is linked to the publication via CrossRef and mentioned accordingly on the website of the respective preprint platform.

After initial skepticism, most publishers now allow papers to be posted on preprint servers prior to submission. An increasing number of journals even allow direct submission of a preprint to their peer review process. The US National Institutes of Health and the Wellcome Trust, among other funders, also encourage prepublication and permit researchers to cite preprints in their grant applications. There are now numerous preprint repositories for different scientific disciplines. BioASAP provides a searchable database for preprint servers that can help in identifying the one that best matches an individual’s needs [ 93 ]. The most popular repository for animal research is bioRxiv, which is hosted by the Cold Spring Harbor Laboratory ( Table 1 ).

The early exchange of scientific results is particularly important for animal research. This acceleration of the publication process can help other scientists to adapt their research or could even prevent animal experiments if other scientists become aware that an experiment has already been done before starting their own. In addition, preprints can help to increase the visibility of research. Journal articles that have a corresponding preprint publication have higher citation and Altmetric counts than articles without preprint [ 94 ]. In addition, the publication of preprints can help to combat publication bias, which represents a major problem in animal research [ 16 ]. Since journals and readers prioritize cutting-edge studies with positive results over inconclusive or negative results, researchers are reluctant to invest time and money in a manuscript that is unlikely to be accepted in a high-impact journal.

In addition to the option of publishing as preprint, other alternative publication formats have recently been introduced to facilitate the publication of research results that are hard to publish in traditional peer-reviewed journals. These include micro publications, data repositories, data journals, publication platforms, and journals that focus on negative or inconclusive results. The tool fiddle can support scientists in choosing the right publication format [ 95 , 96 ].

Open access publication

Publishing open access is one of the most established open science strategies. In contrast to the FAIR data principle, the term open access publication refers usually to the publication of a manuscript on a platform that is accessible free of charge—in translational biomedical research, this is mostly in the form of a scientific journal article. Originally, publications accessible free of charge were the answer to the paywalls established by renowned publishing houses, which led to social inequalities within and outside the research system. In translational biomedical research, the ethical aspect of urgently needed transparency is another argument in favor of open access publication, as these studies will not only be findable, but also internationally readable.

There are different ways of open access publishing; the 2 main routes are gold open access and green open access. Numerous journals offer now gold open access. It refers to the immediate and fully accessible publication of an article. The Directory of Open Access Journals (DOAJ) provides a complete and updated list for high-quality, open access, and peer-reviewed journals [ 97 ]. Charité–Universitätsmedizin Berlin offers a specific tool for biomedical open access journals that supports animal researchers to choose an appropriate journal [ 49 ]. In addition, the Sherpa Romeo platform is a straightforward way to identify publisher open access policies on a journal-by-journal basis, including information on preprints, but also on licensing of articles [ 51 ]. Hybrid open access refers to openly accessible articles in otherwise paywalled journals. By contrast, green open access refers to the publication of a manuscript or article in a repository that is mostly operated by institutions and/or universities. The publication can be exclusively on the repository or in combination with a publisher. In the quality-assured, global Directory of Open Access Repositories (openDOAR), scientists can find thousands of indexed open access repositories [ 49 ]. The publisher often sets an embargo during which the authors cannot make the publication available in the repository, which can restrict the combined model. It is worth mentioning that gold open access is usually more expensive for the authors, as they have to pay an article processing charge. However, the article’s outreach is usually much higher than the outreach of an article in a repository or available exclusively as subscription content [ 98 ]. Diamond open access refers to publications and publication platforms that can be read free of charge by anyone interested and for which no costs are incurred by the authors either. It is the simplest and fairest form of open access for all parties involved, as no one is prevented from participating in scientific discourse by payment barriers. For now, it is not as widespread as the other forms because publishers have to find alternative sources of revenue to cover their costs.

As social media and the researcher’s individual public outreach are becoming increasingly important, it should be remembered that the accessibility of a publication should not be confused with the licensing under which the publication is made available. In order to be able to share and reuse one’s own work in the future, we recommend looking for journals that allow publications under the Creative Commons licenses CC BY or CC BY-NC. This also allows the immediate combination of gold and green open access.

Creative commons licenses

Attributing Creative Commons (CC) licenses to scientific content can make research broadly available and clearly specifies the terms and conditions under which people can reuse and redistribute the intellectual property, namely publications and data, while giving the credit to whom it deserves [ 49 ]. As the laws on copyright vary from country to country and law texts are difficult to understand for outsiders, the CC licenses are designed to be easily understandable and are available in 41 languages. This way, users can easily avoid accidental misuse. The CC initiative developed a tool that enables researchers to find the license that best fits their interests [ 49 ]. Since the licenses are based on a modular concept ranging from relatively unrestricted licenses (CC BY, free to use, credit must be given) to more restricted licenses (CC BY-NC-ND, only free to share for non-commercial purposes, credit must be given), one can find an appropriate license even for the most sensitive content. Publishing under an open CC license will not only make the publication easy to access but can also help to increase its reach. It can stimulate other researchers and the interested public to share this article within their network and to make the best future use of it. Bear in mind that datasets published independently from an article may receive a different CC license. In terms of intellectual property, data are not protected in the same way as articles, which is why the CC initiative in the United Kingdom recommends publishing them under a CC0 (“no rights reserved”) license or the Public Domain Mark. This gives everybody the right to use the data freely. In an animal ethics sense, this is especially important in order to get the most out of data derived from animal experiments.

Data and code repositories

Sharing research data is essential to ensure reproducibility and to facilitate scientific progress. This is particularly true in animal research and the scientific community increasingly recognizes the value of sharing research data. However, even though there is increasing support for the sharing of data, researchers still perceive barriers when it comes to doing so in practice [ 99 – 101 ]. Many universities and research institutions have established research data repositories that provide continuous access to datasets in a trusted environment. Many of these data repositories are tied to specific research areas, geographic regions, or scientific institutions. Due to the growing number and overall heterogeneity of these repositories, it can be difficult for researchers, funding agencies, publishers, and academic institutions to identify appropriate repositories for storing and searching research data.

Recently, several web-based tools have been developed to help in the selection of a suitable repository. One example is Re3data, a global registry of research data repositories that includes repositories from various scientific disciplines. The extensive database can be searched by country, content (e.g., raw data, source code), and scientific discipline [ 49 ]. A similar tool to help find a data archive specific to the field is FAIRsharing, based at Oxford University [ 102 ]. If there is no appropriate subject-specific data repository or one seems unsuitable for the data, there are general data repositories, such as Open Science Framework, figshare, Dryad, or Zenodo. To ensure that data stored in a repository can be found, a DOI is assigned to the data. Choosing the right license for the deposited code and data ensures that authors get credit for their work.

Publication and connection of all outcomes

If scientists have used all available open science tools during the research process, then publishing and linking all outcomes represents the well-deserved harvest ( Fig 2 ). At the end of a research process, researchers will not just have 1 publication in a journal. Instead, they might have a preregistration, a preprint, a publication in a journal, a dataset, and a protocol. Connecting these outcomes in a way that enables other scientists to better assess the results that link these publications will be key. There are many examples of good open science practices in laboratory animal science, but we want to highlight one of them to show how this could be achieved. Blenkuš and colleagues investigated how mild stress-induced hyperthermia can be assessed non-invasively by thermography in mice [ 103 ]. The study was preregistered with animalstudyregistry.org , which is referred to in their publication [ 104 ]. A deviation from the originally preregistered hypothesis was explained in the manuscript and the supplementary material was uploaded to figshare [ 105 ].

Application of open science practices can increase the reproducibility and visibility of a research project at the same time. By publishing different research outputs with more detailed information than can be included in a journal article, researchers enable peers to replicate their work. Reporting according to guidelines and using transparent visualization will further improve this reproducibility. The more research products that are generated, the more credit can be attributed. By communicating on social media or additionally publishing slides from delivered talks or posters, more attention can be raised. Additionally, publishing open access and making the work machine-findable makes it accessible to an even broader number of peers.

https://doi.org/10.1371/journal.pbio.3001810.g002

It might also be helpful to provide all resources from a project in a single repository such as Open Science Framework, which also implements other, different tools that might have been used, like GitHub or protocols.io.

Communicating your research

Once all outcomes of the project are shared, it is time to address the targeted peers. Social media is an important instrument to connect research communities [ 106 ]. In particular, Twitter is an effective way to communicate research findings or related events to peers [ 107 ]. In addition, specialized platforms like ResearchGate can support the exchange of practical experiences ( Table 1 ). When all resources related to a project are kept in one place, sharing this link is a straightforward way to reach out to fellow scientists.

With the increasing number of publications, science communication has become more important in recent years. Transparent science that communicates openly with the public contributes to strengthening society’s trust in research.

Conclusions

Plenty of open science tools are already available and the number of tools is constantly growing. Translational biomedical researchers should seize this opportunity, as it could contribute to a significant improvement in the transparency of research and fulfil their ethical responsibility to maximize the impact of knowledge gained from animal experiments. Over and above this, open science practices also bear important direct benefits for the scientists themselves. Indeed, the implementation of these tools can increase the visibility of research and becomes increasingly important when applying for grants or in recruitment decisions. Already, more and more journals and funders require activities such as data sharing. Several institutions have established open science practices as evaluation criteria alongside publication lists, impact factor, and h-index for panels deciding on hiring or tenure [ 108 ]. For new adopters, it is not necessary to apply all available practices at once. Implementing single tools can be a safe approach to slowly improve the outreach and reproducibility of one’s own research. The more open science products that are generated, the more reproducible the work becomes, but also the more the visibility of a study increases ( Fig 2 ).

As other research fields, such as social sciences, are already a step ahead in the implementation of open science practices, translational biomedicine can profit from their experiences [ 109 ]. We should thus keep in mind that open science comes with some risks that should be minimized early on. Indeed, the more open science practices become incentivized, the more researchers could be tempted to get a transparency quality label that might not be justified. When a study is based on a bad hypothesis or poor statistical planning, this cannot be fixed by preregistration, as prediction alone is not sufficient to validate an interpretation [ 110 ]. Furthermore, a boom of data sharing could disconnect data collectors and analysts, bearing the risk that researchers performing the analysis lack understanding of the data. The publication of datasets could also promote a “parasitic” use of a researcher’s data and lead to scooping of outcomes [ 111 ]. Stakeholders could counteract such a risk by promoting collaboration instead of competition.

During the COVID-19 pandemic, we have seen an explosion of preprint publications. This unseen acceleration of science might be the adequate response to a pandemic; however, the speeding up science in combination with the “publish or perish” culture could come at the expense of the quality of the publication. Nevertheless, a meta-analysis comparing the quality of reporting between preprints and peer-reviewed articles showed that the quality of reporting in preprints in the life sciences is at most slightly lower on average compared to peer-reviewed articles [ 112 ]. Additionally, preprints and social media have shown during this pandemic that a premature and overconfident communication of research results can be overinterpreted by journalists and raise unfounded hopes or fears in patients and relatives [ 113 ]. By being honest and open about the scope and limitations of the study and choosing communication channels carefully, researchers can avoid misinterpretation. It should be noted, however, that by releasing all methodological details and data in research fields such as viral engineering, where a dual use cannot be excluded, open science could increase biosecurity risk. Implementing access-controlled repositories, application programming interfaces, and a biosecurity risk assessment in the planning phase (i.e., by preregistration) could mitigate this threat [ 114 ].

Publishing in open access journals often involves higher publication costs, which makes it more difficult for institutes and universities from low-income countries to publish there [ 115 ]. Equity has been identified as a key aim of open science [ 116 ]. It is vital, therefore, that existing structural inequities in the scientific system are not unintentionally reinforced by open science practices. Early career researchers have been the main drivers of the open science movement in other fields even though they are often in vulnerable positions due to short contracts and hierarchical and strongly networked research environments. Supporting these early career researchers in adopting open science tools could significantly advance this change in research culture [ 117 ]. However, early career researchers can already benefit by publishing registered reports or preprints that can provide a publication much faster than conventional journal publications. Communication in social media can help them establish a network enabling new collaborations or follow-up positions.

Even though open science comes with some risks, the benefits easily overweigh these caveats. If a change towards more transparency is accompanied by the implementation of open science in the teaching curricula of the universities, most of the risks can be minimized [ 118 ]. Interestingly, we have observed that open science tools and infrastructure that are specific to animal research seem to mostly come from Europe. This may be because of strict regulations within Europe for animal experiments or because of a strong research focus in laboratory animal science along with targeted research funding in this region. Whatever the reason might be, it demonstrates the important role of research policy in accelerating the development towards 3Rs and open science.

Overall, it seems inevitable that open science will eventually prevail in translational biomedical research. Scientists should not wait for the slow-moving incentive framework to change their research habits, but should take pioneering roles in adopting open science tools and working towards more collaboration, transparency, and reproducibility.

Acknowledgments

The authors gratefully acknowledge the valuable input and comments from Sebastian Dunst, Daniel Butzke, and Nils Körber that have improved the content of this work.

• View Article
• PubMed/NCBI
• Google Scholar
• 6. Cary Funk MH, Brian Kennedy, Courtney Johnson. Americans say open access to data and independent review inspire more trust in research findings. Pew Research Center Website: Pew Research Center; 2019. Available from: https://www.pewresearch.org/science/2019/08/02/americans-say-open-access-to-data-and-independent-review-inspire-more-trust-in-research-findings/ .
• 7. International Reproducibility Networks. International Networks Statement UK Reproducibility Network Website: UK Reproducibility Network. 2021. Available from: https://cpb-eu-w2.wpmucdn.com/blogs.bristol.ac.uk/dist/b/631/files/2021/09/International-Networks-Statement-v1.0.pdf .
• 13. Article 36 of Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010 amended by Regilation (EU) 2019/1010 of the European Parliament and of the Council of 5 June 2019. OJEU. 2010;L276:36.
• 19. American Association for Cancer Research. Editorial Policies. 2021. Available from: https://aacrjournals.org/content/authors/editorial-policies .
• 21. Diederich K, Schmitt K, Schwedhelm P, Bert B, Heinl C. Open Science Toolbox for Animal Research. Zenodo. 2022. Available from: https://zenodo.org/record/6497560 .
• 29. NC3R. ARRIVE guidelines. NC3R Website. Available from: https://arriveguidelines.org/ .
• 31. Canadian Institutes of Health Research. How to integrate sex and gender into research. Website of the Canadian Institutes of Health Research: Canadian Institutes of Health Research. 2019 [cited 2019 Aug 21]. Available from: https://cihr-irsc.gc.ca/e/50836.html .
• 33. Simon T, Bate RAC. InVivoStat. Available from: https://invivostat.co.uk/ .
• 35. Urbaniak G, Plous S. Research randomizer (version 4.0) [computer software]. 2013.
• 47. Medical Research Council’s. Data sharing policy. UK Research and Innovation Website 2021. Available from: https://www.ukri.org/publications/mrc-data-sharing-policy/ .
• 48. University of California Curation Center. DMPTool. 2021. Available from: https://dmptool.org/ .
• 49. Digital Curation Centre. DMPOnline. Available from: https://dmponline.dcc.ac.uk/ . Digital Curation Centre; 2021.
• 50. Harvard Longwood Medical Area Research Data Management Working Group. Biomedical Data Lifecycle. Harvard Medical School Website: Harvard Medical School; 2021. Available from: https://datamanagement.hms.harvard.edu/about/what-research-data-management/biomedical-data-lifecycle .
• 51. Joint Information Systems Committee. Research data management toolkit JISC Website: JISC; 2018. Available from: https://www.jisc.ac.uk/guides/rdm-toolkit .
• 54. German Centre for the Protection of Laboratory Animals (Bf3R). NTPs—Nicht Technische Projektzusammenfassungen 3R-SMART; 2020. Available from: https://www.3r-smart.de/index.php?id=6895 .
• 55. Understanding Animal Research. Guide to writing non-technical summaries concordat on openness on animal research in the UK2018. Available from: https://concordatopenness.org.uk/guide-to-writing-non-technical-summaries .
• 56. Gerlach B, Untucht C, Stefan A. Electronic Lab Notebooks and Experimental Design Assistants. In: Bespalov A, Michel MC, Steckler T, editors. Good Research Practice in Non-Clinical Pharmacology and Biomedicine. Cham: Springer International Publishing; 2020. p. 257–75.
• 58. Adam BL, Birte L. ELN Guide: electronic laboratory notebooks in the context of research data management and good research practice–a guide for the life sciences. Cologne, Germany: ZB MED–Information Centre for Life Sciences; 2021.
• 59. AgileBio. LabCollector Website https://labcollector.com/labcollector-lims/features/modules/animals-module/2022 . Available from: https://labcollector.com/labcollector-lims/features/modules/animals-module/ .
• 61. Harvard Longwood Medical Area Research Data Management Working Group. Electronic Lab Notebook Comparison Matrix. Zenodo. 2021.
• 70. Bio-protocol. Collaborating Journals bio-protocol website2021. Available from: https://bio-protocol.org/default.aspx?dw=Collaborating .
• 76. Dinkel H. anishare: GitHub; [updated June 2018]. Available from: https://github.com/hdinkel/anishare .
• 89. O’Connor AM. MERIDIAN: Menagerie of Reporting guidelines Involving Animals. Iowa State University; 2022. Available from: https://meridian.cvm.iastate.edu/ .
• 90. The Jackson Laboratory. Mouse Nomenclature Home Page at the Mouse Genome Informatics website World Wide Web: The Jackson Laboratory,Bar Harbor, Maine. Available from: http://www.informatics.jax.org/mgihome/nomen/index.shtml .
• 97. Directory of Open Access Journals. Find open access journals & articles. Available from: https://doaj.org/ . Directory of Open Access Journals, [DOAJ]; 2021.
• 98. Gold Open Access research has greater societal impact as used more outside of academia [press release]. Springer Nature Website: Springer. Nature. 2020;30:2020.
• 104. Franco NH. Can we use infrared thermography for assessing emotional states in mice?—A comparison between handling-induced stress by different techniques. Available from: animalstudyregistry.org . German Federal Institute for Risk Assessment (BfR); 2020. https://doi.org/10.17590/asr.0000224

Ethical care for research animals

WHY ANIMAL RESEARCH?

The use of animals in some forms of biomedical research remains essential to the discovery of the causes, diagnoses, and treatment of disease and suffering in humans and in animals., stanford shares the public's concern for laboratory research animals..

Many people have questions about animal testing ethics and the animal testing debate. We take our responsibility for the ethical treatment of animals in medical research very seriously. At Stanford, we emphasize that the humane care of laboratory animals is essential, both ethically and scientifically.  Poor animal care is not good science. If animals are not well-treated, the science and knowledge they produce is not trustworthy and cannot be replicated, an important hallmark of the scientific method .

There are several reasons why the use of animals is critical for biomedical research: 

••  Animals are biologically very similar to humans. In fact, mice share more than 98% DNA with us!

••  Animals are susceptible to many of the same health problems as humans – cancer, diabetes, heart disease, etc.

••  With a shorter life cycle than humans, animal models can be studied throughout their whole life span and across several generations, a critical element in understanding how a disease processes and how it interacts with a whole, living biological system.

The ethics of animal experimentation

Nothing so far has been discovered that can be a substitute for the complex functions of a living, breathing, whole-organ system with pulmonary and circulatory structures like those in humans. Until such a discovery, animals must continue to play a critical role in helping researchers test potential new drugs and medical treatments for effectiveness and safety, and in identifying any undesired or dangerous side effects, such as infertility, birth defects, liver damage, toxicity, or cancer-causing potential.

U.S. federal laws require that non-human animal research occur to show the safety and efficacy of new treatments before any human research will be allowed to be conducted.  Not only do we humans benefit from this research and testing, but hundreds of drugs and treatments developed for human use are now routinely used in veterinary clinics as well, helping animals live longer, healthier lives.

It is important to stress that 95% of all animals necessary for biomedical research in the United States are rodents – rats and mice especially bred for laboratory use – and that animals are only one part of the larger process of biomedical research.

Our researchers are strong supporters of animal welfare and view their work with animals in biomedical research as a privilege.

Stanford researchers are obligated to ensure the well-being of all animals in their care..

Stanford researchers are obligated to ensure the well-being of animals in their care, in strict adherence to the highest standards, and in accordance with federal and state laws, regulatory guidelines, and humane principles. They are also obligated to continuously update their animal-care practices based on the newest information and findings in the fields of laboratory animal care and husbandry.  

Researchers requesting use of animal models at Stanford must have their research proposals reviewed by a federally mandated committee that includes two independent community members.  It is only with this committee’s approval that research can begin. We at Stanford are dedicated to refining, reducing, and replacing animals in research whenever possible, and to using alternative methods (cell and tissue cultures, computer simulations, etc.) instead of or before animal studies are ever conducted.

Organizations and Resources

There are many outreach and advocacy organizations in the field of biomedical research.

• Learn more about outreach and advocacy organizations

Stanford Discoveries

What are the benefits of using animals in research? Stanford researchers have made many important human and animal life-saving discoveries through their work. 

• Learn more about research discoveries at Stanford

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Recommended Articles
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Toward transparency on animal experimentation in switzerland: seven recommendations for the provision of public information in swiss law.

Simple Summary

1. introduction, 2. what is transparency and why is it important, 2.1. transparency in democracy.

Agent accountability : In this conception, “accountability” denotes the duty owed by an agent to his principal, whereby the principal may demand from the agent an account of the work that the agent has been doing in the principal’s name or on the principal’s behalf, enabling the principal if she sees fit to sanction or replace the agent or terminate the agency relationship [ 19 ].

2.2. Transparency in Animal Experimentation

3. transparency requirements in switzerland, 3.1. documentation and reporting requirements, 3.2. public information, 4. transparency requirements in the european union, 4.1. non-technical project summaries (nts), 4.2. public statistics, 4.3. the german case, 5. toward more transparency in public information under swiss law, 5.1. in experimentation.

• − Numbers of animals used in experiments
• − Species of the animals used in experiments
• − Degrees of severity of harm experienced by the animals used in experiments
• − Objective and subject area of the experiment
• − Title of a research project
• − Fate of the animals at the end of the experiment
• − Costs and funding details of animal experimentation
• − A comprehensible presentation of the harm-benefit analysis.

5.1.1. Fate of the Animals

5.1.2. funding of animal experimentation, 5.1.3. harm-benefit analysis, 5.2. in animal facilities.

• − Number of animals born in facilities
• − Number of animals imported from abroad
• − Number of breeding/surplus animals
• − Fate of breeding/surplus animals
• − Harms experienced by animals in facilities
• − Funding of animal facilities

5.2.1. Number of Breeding/Surplus Animals

5.2.2. fate of breeding/surplus animals, 5.2.3. harms experienced by animals in facilities, 5.2.4. funding of animal facilities, 6. conclusions, author contributions, institutional review board statement, informed consent statement, data availability statement, acknowledgments, conflicts of interest.

• The Federal Council. Popular Initiative “Yes to the Ban on Animal and Human Experiments—Yes to Research That Brings Safety and Progress” of 13 February 2022. Available online: https://www.admin.ch/gov/en/start/documentation/votes/20220213/the-ban-on-animal-and-human-experiments.html (accessed on 6 March 2024).
• European Union. European Citizens’ Initiative (ECI): “SAVE CRUELTY FREE COSMETICS—COMMIT TO A EUROPE WITHOUT ANIMAL TESTING” of 31 August 2022. Available online: https://citizens-initiative.europa.eu/initiatives/details/2021/000006_en (accessed on 6 March 2024).
• European Union. European Citizens’ Initiative (ECI): “Stop Vivisection” of 3 March 2015. Available online: https://europa.eu/citizens-initiative/initiatives/details/2012/000007_de (accessed on 6 March 2024).
• Persson, K.; Rodriguez Perez, C.; Louis-Maerten, E.; Müller, N.; Shaw, D. Killing in the name of 3R? The ethics of death in animal research. JAGE 2024 . under revision. [ Google Scholar ]
• Lambert, H.; Elwin, A.; D’Cruze, N. Wouldn’t hurt a fly? A review of insect cognition and sentience in relation to their use as food and feed. Appl. Anim. Behav. Sci. 2021 , 243 , 105432. [ Google Scholar ] [ CrossRef ]
• Swiss Universities. Swiss Transparency Agreement on Animal Research (STAAR). Available online: https://www.swissuniversities.ch/organisation/gremien/kammer-universitaere-hochschulen/swiss-transparency-agreement-on-animal-research-staar (accessed on 6 March 2024).
• Peters, A. Towards Transparency as a Global Norm. In Transparency in International Law ; Bianchi, A., Peters, A., Eds.; Cambridge University Press: Cambridge, UK, 2013; pp. 534–607. ISBN 978-113-910-88-43. [ Google Scholar ]
• Hollyer, J.R.; Rosendorff, B.P.; Vreeland, J.R. Democracy and Transparency. J. Politics 2011 , 73 , 1191–1205. [ Google Scholar ] [ CrossRef ]
• Tschannen, P. Staatsrecht der Schweizerischen Eidgenossenschaft , 5th ed.; Stämpfli: Bern, Switzerland, 2011. [ Google Scholar ]
• Rhinow, R.; Schefer, M.; Uebersax, P. Schweizerisches Verfassungsrecht , 3rd ed.; Helbing Lichtenhahn: Basel, Switzerland, 2016. [ Google Scholar ]
• Wiederkehr, R. Transparenz als Grundsatz rechtsstaatlichen Handelns (Art. 5 BV). Schweiz. Zentralblatt Staats- Verwaltungsrecht 2007 , 108 , 521–543. [ Google Scholar ]
• Wiederkehr, R.; Meyer, C.; Böhme, A. Transparenz als Verfahrensgrundsatz: Gesetzliche Transparenztechniken und praktische Argumentationsfelder. Recht 2022 , 131–148. [ Google Scholar ]
• Sprecher, F. Transparenz—Ein Grundprinzip des Rechtsstaats und seine Bedeutung im Gesundheitsrecht, insbesondere im Heilmittel- und Humanforschungsrecht. Z. Schweiz. Recht 2016 , 135 , 143–241. [ Google Scholar ]
• European Union. Transparency Register. Available online: https://commission.europa.eu/about-european-commission/service-standards-and-principles/transparency/transparency-register_en (accessed on 27 March 2024).
• Decision by the Swiss Federal Supreme Court BGE 145 I 282, Consideration 4.1. Available online: https://bger.li/145-I-282 (accessed on 18 July 2024).
• Decision by the Swiss Federal Supreme Court BGE 114 Ia 427, Consideration 5c; BGE 143 I 78, Consideration 4.4. Available online: https://bger.li/114-Ia-427 (accessed on 18 July 2024).
• Kogelmann, B. Secrecy and transparency in political philosophy. Philos. Compass 2021 , 16 , e12733. [ Google Scholar ] [ CrossRef ]
• Brunner, S. Offentlichkeit der Verwaltung und informationelle Selbstbestimmung: Von Kollisionen und Verkehrsregeln. In Selbstbestimmung und Recht, Festgabe für Rainer J. Schweizer zum 60. Geburtstag ; Sutter, P., Ed.; Schulthess: Zurich, Switzerland, 2003. [ Google Scholar ]
• Waldron, J. Accountability and Insolence. In Political Theory: Essays on Institutions ; Harvard University Press: Cambridge, MA, USA; London, UK, 2016; pp. 167–194. [ Google Scholar ]
• Merkes, M.; Buttrose, R. Increasing the Transparency of Animal Experimentation: An Australian Perspective. In Animal Experimentation: Working towards a Paradigm Shift ; Herrmann, K., Jayne, K., Eds.; Brill: Leiden, Netherlands, 2019; pp. 224–243. [ Google Scholar ]
• McLeod, C.; Hobson-West, P. Opening up animal research and science–society relations? A thematic analysis of transparency discourses in the United Kingdom. Public Underst. Sci. 2016 , 7 , 791–806. [ Google Scholar ] [ CrossRef ]
• O’Sullivan, S. Transparency and Animal Regulation: An Australian Case Study. Anim. Lib. Philos. Policy J. 2006 , 4 , 1–25. [ Google Scholar ]
• Animal Research Tomorrow. Basel Declaration “A Call for more Trust, Transparency and Communication on Animal Research” of 29 November 2010. Available online: https://animalresearchtomorrow.org/en/basel-declaration (accessed on 6 March 2024).
• Animal Rights Switzerland. Tierversuche: Alternativen Mehr Fördern. Available online: https://animal-rights-switzerland.ch/themen-tierversuche-in-der-schweiz/ (accessed on 25 March 2024).
• Stiftung für das Tier im Recht. Tierversuchsstatistik 2018—TIR Fordert Mehr Transparenz. Available online: https://www.tierimrecht.org/de/news/newsmeldungen-2019/2019-07-15-tierversuchsstatistik-2018--tir-fordert-mehr-transparenz/ (accessed on 25 March 2024).
• Interpellation Schneider (22.3808) “Aussagekräftige und Transparente Tierversuchsstatistik” of 17 June 2022. Available online: https://www.parlament.ch/de/ratsbetrieb/suche-curia-vista/geschaeft?AffairId=20223808 (accessed on 27 March 2024).
• Postulate Graf (22.3612) “Wie Kann das mit Grossem Tierleid Behaftete Züchten und Töten Hunderttausender Labortiere Reduziert Werden?” of 14 June 2022. Available online: https://www.parlament.ch/de/ratsbetrieb/suche-curia-vista/geschaeft?AffairId=20223612 (accessed on 27 March 2024).
• Request Graf (11.1085) “Öffentliche Gelder für Tierversuche. Mehr Transparenz” of 30 September 2011. Available online: https://www.parlament.ch/de/ratsbetrieb/suche-curia-vista/geschaeft?AffairId=20111085 (accessed on 27 March 2024).
• Doctors Against Animal Experiments. Survey: Majority against Animal Testing! Available online: https://www.aerzte-gegen-tierversuche.de/en/basic-infos/animal-experiments/statistics/survey-majority-against-animal-testing (accessed on 27 March 2024).
• Ipsos MORI. Openness in Animal Research. The Public’s Views on Openness and Transparency in Animal Research . Available online: https://www.ipsos.com/sites/default/files/publication/1970-01/sri-health-openness-in-animal-research-2013.pdf (accessed on 27 March 2024).
• Doctors against Animal Experiments. Results of Questionnaire for the General Public on the Revision of Directive 86/609/EEC on the Protection of Animals Used for Experimental and Other Scientific Purposes. Available online: https://www.aerzte-gegen-tierversuche.de/images/themen/statistiken/umfragen/eu_umfrage_2006.pdf (accessed on 23 March 2024).
• Hadley, J. Telling it Like it is: A Proposal to Improve Transparency in Biomedical Research. Between Species 2012 , 15 , 103–126. [ Google Scholar ] [ CrossRef ]
• Stephanie, C.; Tim, P. The ethics of transparency. In The Routledge Companion to Journalism Ethics ; Routledge: London, UK, 2021; pp. 175–183. [ Google Scholar ]
• Kiani, A.; Pheby, D.; Henehan, G.; Brown, R.; Sieving, P.; Sykora, P.; Marks, R.; Falsini, B.; Capodicasa, N.; Miertus, S.; et al. Ethical considerations regarding animal experimentation. JPMH 2022 , 63 , E255–E266. [ Google Scholar ] [ PubMed ]
• Eurogroup for Animals. 77% of EU Citizens Want a Transition to Non-Animal Science. Available online: https://www.eurogroupforanimals.org/news/77-eu-citizens-want-transition-non-animal-science (accessed on 27 March 2024).
• Swiss Federal Audit Office. Tierversuche. Prüfung des Bewilligungsprozesses, der Kosten und der Finanzierung. Available online: https://www.efk.admin.ch/en/audit/animal-experiments-approval-process-costs-and-financing/ (accessed on 27 March 2024).
• Gerritsen, V. Der gesellschaftliche Nutzen von Tierversuchen. TIERethik 2024 , 28 , 61–90. [ Google Scholar ]
• Federal Council. Botschaft zur Änderung des Tierschutzgesetzes of 7 September 2011, BBl 2011 7055. Available online: https://www.fedlex.admin.ch/eli/fga/2011/1123/de (accessed on 18 July 2024).
• Stiglitz, J.E. Democratizing the International Monetary Fund and the World Bank: Governance and Accountability. Gov. Int. J. Policy Adm. Inst. 2003 , 16 , 111–139. [ Google Scholar ] [ CrossRef ]
• Carbone, L. Open Transparent Communication about Animals in Laboratories: Dialog for Multiple Voices and Multiple Audiences. Animals 2021 , 11 , 368. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• SAMS; SCNAT. Ethsiche Grundsätze und Richtlinien für Tierversuche. Available online: https://www.samw.ch/dam/jcr:8d5c661c-9988-4f9b-89d4-1d92cdf365d1/samw-scnat_ethische-grundsaetze-und-richtlinien-fuer-tierversuche_2005.pdf (accessed on 27 March 2024).
• Gehrig, T.K. Struktur und Instrumente des Tierschutzrechts ; Schulthess: Zurich, Switzerland, 1990. [ Google Scholar ]
• Swiss Confederation. Federal Food Safety and Veterinary Office. Statistics. Available online: https://www.tv-statistik.ch/ (accessed on 6 March 2024).
• Tschannen, P.; Müller, M.; Kern, M. Allgemeines Verwaltungsrecht , 5th ed.; Stämpfli: Bern, Switzerland, 2022. [ Google Scholar ]
• Häfelin, U.; Müller, G.; Uhlmann, F. Allgemeins Verwaltungsrecht , 8th ed.; Dike: Zurich/St. Gallen, Switzerland, 2020. [ Google Scholar ]
• Federal Council. Erläuterungen zur Totalrevision der Tierschutzverordnung (TschV; SR 455.1). Available online: https://www.fedlex.admin.ch/filestore/fedlex.data.admin.ch/eli/dl/proj/6005/106/cons_1/doc_1/de/pdf-a/fedlex-data-admin-ch-eli-dl-proj-6005-106-cons_1-doc_1-de-pdf-a.pdf (accessed on 27 March 2024).
• Consultation Drafts of the AniPO and the AEO. Available online: https://www.fedlex.admin.ch/de/consultation-procedures/ongoing#https://fedlex.data.admin.ch/eli/dl/proj/2022/56/cons_1 (accessed on 27 March 2024).
• The Federal Council. Press Release of the Swiss Federal Council “Verbesserung des Tierwohls und Striktere Regeln für den Hundeimport: Vernehmlassung eröffnet” of 27 November 2023. Available online: https://www.admin.ch/gov/de/start/dokumentation/medienmitteilungen.msg-id-98936.html (accessed on 6 March 2024).
• Available online: https://www.blv.admin.ch/blv/en/home/tiere/tierversuche/forschende/animex-ch.html (accessed on 27 March 2024).
• Federal Food Safety and Veterinary Office. Bericht Tierversuchsstatistik 2022. Available online: https://www.blv.admin.ch/blv/de/home/tiere/tierversuche/bericht-tierversuchsstatistik.html (accessed on 27 March 2024).
• STS Rehoming Project. Available online: https://adopt-a-pet.ch/rehoming-projekt/ (accessed on 27 March 2024).
• Döring, D.; Erhard, M.H. Verbleib von überzähligen und überlebenden Versuchstieren. ALTEX 2005 , 22 , 7–11. [ Google Scholar ]
• DeGrazia, D. Taking Animals Seriously: Mental Life and Moral Status , 1st ed.; Cambridge University Press: Cambridge, UK; New York, NY, USA; Melbourne, Australia, 1996. [ Google Scholar ]
• Gerritsen, V. Güterabwägung im Tierversuchsbewilligungsverfahren ; Schulthess: Zurich, Switzerland, 2022. [ Google Scholar ]
• Federal Food Safety and Veterinary Office. Publikation der Abgeschlossenene Versuche. Available online: https://www.tv-statistik.ch/de/abgeschlossene-versuche/ (accessed on 27 March 2024).
• Gupta, A. Transparency Under Scrutiny: Information Disclosure in Global Environmental Governance. Glob. Environ. Politics 2008 , 8 , 1–7. [ Google Scholar ] [ CrossRef ]
• Commission to the European Parliament and the Council. Report on the Implementation of Directive 2010/63/EU on the Protection of Animals Used for Scientific Purposes in the Member States of the European Union, 5 February 2020, COM(2020) 15 Final. Available online: https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52020DC0015&rid=9 (accessed on 27 March 2024).
• European Commission. Commission to the European Parliament, The Council, The European Economic and Social Committee and the Committee of the Regions. Report in Accordance with Article 58 of Directive 2010/63/EU on the Protection of Animals Used for Scientific Purposes, COM(2017) 631 Final, 2. Available online: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=SWD:2017:353:REV1&from=EN (accessed on 27 March 2024).
• Council of the European Union. Commission Staff Working Paper Accompanying the Proposal for a Directive of the European Parliament and the Council on the Protection of Animals Used for Scientific Purposes, SEC(2008) 2410 of 5 November 2008. Available online: https://data.consilium.europa.eu/doc/document/ST-15546-2008-ADD-1/en/pdf (accessed on 27 March 2024).
• European Commission, Directorate-General for Environment. Caring for Animals Aiming for Better Science—Directive 2010/63/EU on Protection of Animals Used for Scientific Purposes—Project Evaluation and Retrospective Assessment ; Publications Office of the European Union: Luxembourg, 2018. [ Google Scholar ]
• Annex I of the Commission Implementing Decision (EU) 2020/569 of 16 April 2020 Establishing a Common Format and Information Content for the Submission of the Information to be Reported by the Member States Pursuant to Directive 2010/63/EU of the European Parliament and of the Council on the Protection of Animals Used for Scientific Purposes Repealing Commission Implementing Decision 2012/707/EU. Available online: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32020D0569 (accessed on 27 March 2024).
• European Commission. ALURES Non-Technical Summary EU Database. Available online: https://webgate.ec.europa.eu/envdataportal/web/resources/alures/submission/nts/list (accessed on 27 March 2024).
• European Commission. ALURES Statistical EU Database. Available online: https://webgate.ec.europa.eu/envdataportal/content/alures/section1_number-of-animals.html (accessed on 7 July 2024).
• Rodriguez Perez, C.; Müller, N.D.; Persson, K.; Shaw, D.M. The Ranking Argument—Challenging Favourable Comparative Rhetoric about Animal Welfare Law. LEOH J. Anim. Law Ethics One Health 2023 , 44–55. [ Google Scholar ] [ CrossRef ]
• Animal Study Registry. Available online: https://www.animalstudyregistry.org/asr_web/index.action (accessed on 27 March 2024).
• Bundesinstitut für Risikobewertung. Transparenz Fördern, um Tierversuche zu Vermeiden. Available online: https://www.bfr.bund.de/de/presseinformation/2021/29/transparenz_foerdern__um_tierversuche_zu_vermeiden-276955.html (accessed on 27 March 2024).
• BF3R. Deutsches Zentrum zum Schutz von Verscuhstieren. Verwendung von Versuchstieren im Jahr 2021 . Available online: https://www.bf3r.de/de/verwendung_von_versuchstieren_im_jahr_2021-309160.html (accessed on 27 March 2024).
• Deutsche Juristische Gesellschaft für Tierschutz. Gemeinsame Pressemitteilung der Verbände Ärzte gegen Tierversuche und DJGT zur Tötung von Überschusstieren. Available online: https://djgt.de/2022/05/12/gemeinsame-pressemitteilung-der-verbaende-aerzte-gegen-tierversuche-und-djgt-zur-toetung-von-ueberschusstieren/ (accessed on 27 March 2024).
• Judgement of the German Federal Administrative Court 3 C 28.16 of 13 June 2019. Available online: https://www.bverwg.de/en/130619U3C28.16.0 (accessed on 18 July 2024).
• Feldwisch-Drentrup, H. Germany weighs whether culling excess lab animals is a crime. Science 2022 , 376 , 567–568. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Bf3R. Verwendung von Versuchstieren im Berichtsjahr 2022. Available online: https://www.bf3r.de/de/verwendung_von_versuchstieren_im_berichtsjahr_2022-313306.html (accessed on 5 June 2024).
• Wagenknecht, T.; Eusemann, B.; Schwedhelm, P.; Schönfelder, G.; Bert, B. Das Vorliegen eines “vernünftigen Grundes” bei der Tötung überzähliger Versuchstiere. NuR 2023 , 45 , 225–233. [ Google Scholar ] [ CrossRef ]
• Bundesinstitut für Risikobewertung. Versuchstierschutz: Die Tötung eines “überzähligen” Tieres Ist im Einzelfall Sorgfältig Abzuwägen. Available online: https://www.bf3r.de/cm/343/versuchstierschutz-die-toetung-eines-ueberzaehligen-tieres-ist-im-einzelfall-sorgfaeltig-abzuwaegen.pdf (accessed on 27 March 2024).
• Wewetzer, H.; Wagenknecht, B.; Bert, B. The fate of surplus laboratory animals: Minimizing the production of surplus animals has greatest potential to reduce the number of laboratory animals. EMBO Rep. 2023 , 24 , e56551. [ Google Scholar ] [ CrossRef ]
• Regan, T. The Case for Animal Rights , 2nd ed.; University of California Press: Berkeley, CA, USA, 2004. [ Google Scholar ]
• McMahan, J. The Comparative Badness for Animals of Suffering and Death. In The Ethics of Killing Animals ; Višak, T., Garner, R., Eds.; Oxford University Press: Oxford, UK; New York, NY, USA, 2016; pp. 65–85. [ Google Scholar ]
• Singer, P. Practical Ethics , 3rd ed.; Cambridge University Press: New York, NY, USA, 2011. [ Google Scholar ]
• Torella, K.; Save the Male Chicks. Vox. 2023. Available online: https://www.vox.com/future-perfect/2023/5/1/23700952/egg-industry-male-chick-culling-animal-welfare (accessed on 6 March 2024).
• Vinci, C. European Parliamentary Research Service (EPRS). Male Chick Culling. Available online: https://www.europarl.europa.eu/RegData/etudes/ATAG/2022/739246/EPRS_ATA(2022)739246_EN.pdf (accessed on 6 March 2024).
• Clemence, M.; Leaman, J. Public Attitudes to Animal Research in 2016 ; Department for Business, Energy & Industrial Strategy: London, UK; Ipsos MORI Social Research Institute: London, UK, 2016; Available online: https://doc.ukdataservice.ac.uk/doc/8059/mrdoc/pdf/8059_ols_public_attitudes_to_animal_research_report.pdf (accessed on 6 March 2024).
• Persson, K.; Rodriguez Perez, C.; Hartstang, S.; Kunzmann, P.; Hiebl, B.; Shaw, D. What and why is the “evolutionary scale” in Swiss animal welfare law? LEOH 2024 . submitted. [ Google Scholar ]
• Gremmen, B.; Blok, V. The lesser of two evils? The killing of day-old male chicks in the Dutch egg sector. In Food Futures: Ethics, Science and Culture ; Wageningen Academic Publishers: Wageningen, The Netherlands, 2016; pp. 72–75. [ Google Scholar ]
• Gremmen, B.; Bruijnis, M.R.N.; Blok, V.; Stassen, E.N. A Public Survey on Handling Male Chicks in the Dutch Egg Sector. J. Agric. Environ. Ethics 2018 , 31 , 93–107. [ Google Scholar ] [ CrossRef ]
• Gunasekera, C.A. The Ethics of Killing ‘Surplus’ Zoo Animals. J. Anim. Ethics 2018 , 8 , 93–102. [ Google Scholar ] [ CrossRef ]
• Vuilleumier, M. Les Poussins ne Seront Plus Broyés Vivants en Suisse. SWI. 2019. Available online: https://www.swissinfo.ch/fre/societe/protection-des-animaux_les-poussins-ne-seront-plus-broy%C3%A9s-vivants-en-suisse/45238138 (accessed on 6 March 2024).
• Gerritsen, V.; Rüttimann, A. Neue Wege im Tierversuchsrecht. In Animal Law—Tier und Recht. Entwicklungen und Perspektiven im 21. Jahrhundert ; Michel, M., Kühne, D., Hänni, J., Eds.; Dike: Zurich/St. Gallen, Switzerland, 2012. [ Google Scholar ]
• Herrmann, K. Refinement on the way towards replacement: Are we doing what we can? In Animal Experimentation: Working towards a Paradigm Change ; Hermann, K., Jayne, K., Eds.; Brill: Leiden, The Netherlands, 2019; pp. 1–64. [ Google Scholar ]
• Akhtar, A. Flaws and human harms of animal experimentation. Camb. Q. Healthc. Ethics 2015 , 24 , 407–419. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Hutchinson, I.; Owen, C.; Bailey, J. Modernizing Medical Research to Benefit People and Animals. Animals 2022 , 12 , 1173. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Eggel, M.; Grimm, H. The logic, methodological and practical flaws of the HarmBenefit-Analysis (HBA) in Directive 2010/63/EU. In EurSafe 2018 ; Wageningen Academic Publishers: Vienna, Austria, 2018; pp. 407–411. [ Google Scholar ]
• Federal Food Safety and Veterinary Office. Fachinformation Tierversuche. Schweregrade 1.04. Available online: https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.blv.admin.ch/dam/blv/de/dokumente/tiere/publikationen-und-forschung/tierversuche/klassifikation-schweregrad-tv.pdf.download.pdf/116104_DE.pdf&ved=2ahUKEwjYo9elgZWFAxV1hv0HHeD3AP8QFnoECA4QAQ&usg=AOvVaw31YYmYI8MFQfB2gwrXdWAH (accessed on 27 March 2024).
• Administrative Court of the Canton of Zurich, Decision of 24 November 2022, VB.2021.00276. Available online: https://entscheidsuche.ch/view/ZH_VG_001_-VB-2021-00276_2022-11-24 (accessed on 18 July 2024).
• Holmberg, T.; Ideland, M. Secrets and lies: “selective openness” in the apparatus of animal experimentation. Public Underst. Sci. 2012 , 3 , 354–368. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• KOFAM. Koordinationsstelle Forschung am Menschen. Available online: https://kofam.ch/de (accessed on 27 March 2024).
• Bailey, J. Does the stress inherent to laboratory life and experimentation on animals adversely affect research data? Altern. Lab. Anim. ATLA 2017 , 45 , 299. [ Google Scholar ] [ CrossRef ]
• Lee, G.H.; Kim, K.; Jo, W. Stress Evaluation of Mouse Husbandry Environments for Improving Laboratory Animal Welfare. Animals 2023 , 13 , 249. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Cait, J.; Cait, A.; Scott, R.W.; Winder, C.B.; Mason, G.J. Conventional laboratory housing increases morbidity and mortality in research rodents: Results of a meta-analysis. BMC Biol. 2022 , 20 , 15. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Von Kürthy, C. Tierversuche: Die Qual der Stellvertreter. Fachgespräch “Tierschutz, Tierversuche, Labortierhaltungen” of 7 November 2013. Available online: https://docplayer.org/67834063-Tierversuche-die-qual-der-stellvertreter.html (accessed on 17 April 2024).
• Rodriguez Perez, C.; Persson, K.; Cajiga Morales, R.M.; Elger, B.S.; Shaw, D.M. Russell and Burch’s 3Rs then and now: The case of Switzerland. ALTEX Altern. Anim. Exp. 2023 , 40 , 635–648. [ Google Scholar ]

Share and Cite

Lüthi, N.; Rodriguez Perez, C.; Persson, K.; Elger, B.S.; Shaw, D. Toward Transparency on Animal Experimentation in Switzerland: Seven Recommendations for the Provision of Public Information in Swiss Law. Animals 2024 , 14 , 2154. https://doi.org/10.3390/ani14152154

Lüthi N, Rodriguez Perez C, Persson K, Elger BS, Shaw D. Toward Transparency on Animal Experimentation in Switzerland: Seven Recommendations for the Provision of Public Information in Swiss Law. Animals . 2024; 14(15):2154. https://doi.org/10.3390/ani14152154

Lüthi, Nicole, Christian Rodriguez Perez, Kirsten Persson, Bernice Simone Elger, and David Shaw. 2024. "Toward Transparency on Animal Experimentation in Switzerland: Seven Recommendations for the Provision of Public Information in Swiss Law" Animals 14, no. 15: 2154. https://doi.org/10.3390/ani14152154

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

• Help Animals
• Investigations
• Animal Rights Issues
• Students Opposing Speciesism (SOS)
• Join the Action Team
• Monthly Giving
• Your Legacy For Animals
• PETA Vanguard Society
• ‘In Honor of’ and Memorial Gifts
• Membership Services
• More Ways to Support PETA
• Report Cruelty to Animals
• Urgent Alerts
• Our Campaigns
• Action Team
• Activist Guide
• Get Active Online
• Leaflets & Stickers
• Adoptable Animals
• Rescue Stories
• Recent Investigations
• Investigations & Rescue Fund

Animals Used for Experimentation

• Entertainment
• Companion Animals
• Personal Care & Fashion
• Food & Health
• Humane Home
• Animal Companions
• Students Opposing Speciesism
• Cruelty-Free Database
• PETA Shopping Guide
• PETA Approved Vegan
• PETA Presents
• Work at PETA
• PETA Global
• Vanguard Society
• More Ways to Support
• Gifts in Wills

Right now, millions of mice, rats, rabbits, primates, cats, dogs, and other animals are locked inside cages in laboratories across the country. They languish in pain, ache with loneliness, and are deprived of everything that’s natural and important to them. All they can do is sit and wait in fear of the next terrifying and painful procedure that will be performed on them. A lack of environmental enrichment and the stress of their living situations cause some animals to develop neurotic behaviors, such as incessantly spinning in circles, rocking back and forth, pulling out their own fur, and even biting themselves. After enduring lives of pain, loneliness, and terror, almost all of them will be killed.

How PETA Helps Animals in Laboratories

Since PETA’s inception and the landmark  Silver Spring monkeys  case, we’ve been at the forefront of exposing and ending experiments on animals. Our scientists, campaigners, researchers, and other dedicated staff work hard to persuade universities, hospitals,  contract laboratories ,  other companies , and government agencies to abandon animal tests and embrace modern, non-animal methods.

Two teams lead PETA’s efforts to end tests on animals. Our Laboratory Investigations Department focuses on ending the use of animals in experiments not required by law, and our Regulatory Toxicology Department focuses on replacing the use of animals in tests required by law with human-relevant, animal-free toxicity testing approaches. With help from supporters like you, these teams and other hardworking staff at PETA win numerous  victories  for animals imprisoned in laboratories every year. Here’s how they do it:

• Promoting PETA’s Research Modernization Deal , the first comprehensive, science-backed plan to phase out tests on animals
• Conducting groundbreaking  eyewitness investigations and colorful advocacy campaigns to shut down laboratories and areas of animal experimentation
• Filing groundbreaking lawsuits to challenge public funding of wasteful, cruel animal experiments
• Working with members of Congress to enact laws to replace animals in laboratories
• Persuading government agencies to stop conducting and  requiring experiments on animals
• Encouraging  pharmaceutical, chemical , and  consumer product companies to replace tests on animals with more effective, non-animal methods
• Ending the use of animals in experiments at colleges and universities
• Helping  students and  teachers  end animal dissection in the classroom
• Developing and funding humane non-animal research methods
• Publishing scientific papers on reliable non-animal test methods and presenting them at scientific conferences
• Hosting free workshops and online seminars to share information about animal-free toxicity testing methods
• Urging  health charities not to invest in dead-end tests on animals

How Animals Are Exploited in Laboratories

More than 110 million animals suffer and die in the U.S. every year in cruel chemical, drug, food, and cosmetics tests. They also experience this fate in  medical training exercises , curiosity-driven  experiments at universities ,  classroom biology experiments , and  dissection even though modern, non-animal methods have repeatedly been shown to have more educational value, save teachers time, and save schools money. Exact numbers aren’t available, because mice, rats, birds, and cold-blooded animals—who make up more than 99% of animals used in experiments—aren’t covered by even the minimal protections of the federal Animal Welfare Act and therefore go uncounted.

Examples of chemical and toxicity tests on animals include forcing mice and rats to inhale toxic fumes, force-feeding dogs chemicals, and applying corrosive chemicals into rabbits’ sensitive eyes. Even if a product harms animals, it can still be marketed to consumers. Conversely, just because a product was shown to be safe in animals doesn’t guarantee that it will be safe to use in humans.

Much product testing conducted on animals today isn’t required by law. In fact, a number of countries have implemented bans on the testing of certain types of consumer goods on animals, such as the cosmetics testing bans in India, Israel, New Zealand, Norway, and elsewhere.

Meanwhile, at universities and other institutions, experimenters inflict suffering on and kill animals for little more than curiosity’s sake—even though the vast majority of their findings fail to advance human health . They tear baby monkeys away from their mothers , sew kittens’ eyes shut , mutilate owls’ brains , puncture the intestines of mice so that feces leak into their stomachs , and terrorize songbirds with the sounds of predators . At the end of experiments like these—which consume billions in taxpayer funds and charitable donations each year—almost all the animals are killed.

Animal Experiments Throughout History: A Century of Suffering

PETA created an interactive timeline, “ Without Consent ,” featuring almost 200 stories of animal experiments from the past century to open people’s eyes to the long history of suffering inflicted on nonconsenting animals in laboratories and to challenge them to rethink this exploitation. Visit “ Without Consent ” to learn more about harrowing animal experiments throughout history and how you can help create a better future for living, feeling beings.

Advancing Science Without Suffering: Animal-Free Testing

Testing on animals has been a spectacular failure that has resulted in the loss of trillions of dollars and has cost the lives of innumerable humans and other animals. Experiments on one species frequently fail to predict results in another. Even the National Institutes of Health, the world’s largest funder of biomedical research, acknowledges that 95% of all drugs that are shown to be safe and effective in animal tests fail in human trials.

Technologically advanced  non-animal research methods —such as those using human cells, computational models, or clinical studies—can be used in place of animal testing. These methods are more humane, have the potential to be faster, and are more relevant to humans.

Scientists in PETA’s Science Advancement & Outreach division , a part of the Laboratory Investigations Department, have developed a roadmap to phase out failing tests on animals with sophisticated, animal-free methods. Their Research Modernization Deal has gained the support of scientists, medical doctors, members of Congress, and thousands of others who care about ethical and effective science.

How You Can Help Animals Used in Experiments

Each of us can help prevent the suffering and deaths of animals in laboratories. Here are a few easy ways to get started:

• Sign up for PETA’s Action Team to be alerted when protests are taking place in your area.
• Urge your members of Congress to support PETA’s Research Modernization Deal .
• Search PETA’s Beauty Without Bunnies database to ensure that you’re buying only cruelty-free products.
• Donate only to charities that don’t experiment on animals .
• Request alternatives to animal dissection at your school.
• Call on your alma mater to stop experimenting on animals.
• Share information about animal experimentation issues with your friends and family—and invite them to join you in speaking up for animals.

Popular Resources

PETA’s Beauty Without Bunnies program is the ultimate resource for conscientious shoppers, making the quest for cruelty-free products as easy as pie.

Stand up for your rights in five easy steps!

Take PETA’s Shopping Guide along with you next time you head to the store! Our handy guide will help you find humane products at a glance.

Sign the pledge and joining the thousands of caring consumers who refuse to purchase products from corporations that abuse animals.

By submitting this form, you’re acknowledging that you have read and agree to our privacy policy and agree to receive e-mails from us.

“Almost all of us grew up eating meat, wearing leather, and going to circuses and zoos. We never considered the impact of these actions on the animals involved. For whatever reason, you are now asking the question: Why should animals have rights?” READ MORE

— Ingrid E. Newkirk, PETA President and co-author of Animalkind

Frankie the capuchin is kept in a cage all alone in the middle of the retail store Pet Stop. Text FRANKIE to 73822 to urge Pet Stop to send Frankie to a sanctuary!

Terms for automated texts/calls from PETA: http://peta.vg/txt . Text STOP to end, HELP for more info. Msg/data rates may apply. U.S. only.

Watch CBS News

Here's what a Sam Altman-backed basic income experiment found

By Megan Cerullo

Edited By Anne Marie Lee

Updated on: July 23, 2024 / 10:33 AM EDT / CBS News

A recent study on basic income, backed by OpenAI founder Sam Altman, shows that giving low-income people guaranteed paydays with no strings attached can lead to their working slightly less, affording them more leisure time. 

The study, which is one of the largest and most comprehensive of its kind, examined the impact of guaranteed income on recipients' health, spending, employment, ability to relocate and other facets of their lives.

Altman first announced his desire to fund the study in a 2016 blog post on startup accelerator Y Combinator's site.

Some of the questions he set out to answer about how people behave when they're given free cash included, "Do people sit around and play video games, or do they create new things? Are people happy and fulfilled?" according to the post. Altman, whose OpenAI is behind generative text tool ChatGPT, which threatens to take away some jobs, said in the blog post that he thinks technology's elimination of "traditional jobs"  could make universal basic income necessary in the future. 

How much cash did participants get?

For OpenResearch's Unconditional Cash Study , 3,000 participants in Illinois and Texas received $1,000 monthly for three years beginning in 2020. The cash transfers represented a 40% boost in recipients' incomes. The cash recipients were within 300% of the federal poverty level, with average incomes of less than $29,000. A control group of 2,000 participants received $50 a month for their contributions.

Basic income recipients spent more money, the study found, with their extra dollars going toward essentials like rent, transportation and food.

Researchers also studied the free money's effect on how much recipients worked, and in what types of jobs. They found that recipients of the cash transfers worked 1.3 to 1.4 hours less each week compared with the control group. Instead of working during those hours, recipients used them for leisure time. 

"We observed moderate decreases in labor supply," Eva Vivalt, assistant professor of economics at the University of Toronto and one of the study's principal investigators, told CBS MoneyWatch. "From an economist's point of view, it's a moderate effect." 

More autonomy, better health

Vivalt doesn't view the dip in hours spent working as a negative outcome of the experiment, either. On the contrary, according to Vivalt. "People are doing more stuff, and if the results say people value having more leisure time — that this is what increases their well-being — that's positive." 

In other words, the cash transfers gave recipients more autonomy over how they spent their time, according to Vivalt. 

"It gives people the choice to make their own decisions about what they want to do. In that sense, it necessarily improves their well-being," she said. 

Researchers expected that participants would ultimately earn higher wages by taking on better-paid work, but that scenario didn't pan out. "They thought that if you can search longer for work because you have more of a cushion, you can afford to wait for better jobs, or maybe you quit bad jobs," Vivalt said. "But we don't find any effects on the quality of employment whatsoever."

Uptick in hospitalizations

At a time when even Americans with insurance say they have trouble staying healthy because they struggle to afford care , the study results show that basic-income recipients actually increased their spending on health care services. 

Cash transfer recipients experienced a 26% increase in the number of hospitalizations in the last year, compared with the average control recipient. The average recipient also experienced a 10% increase in the probability of having visited an emergency department in the last year.

Researchers say they will continue to study outcomes of the experiment, as other cities across the U.S. conduct their own tests of the concept.

Megan Cerullo is a New York-based reporter for CBS MoneyWatch covering small business, workplace, health care, consumer spending and personal finance topics. She regularly appears on CBS News 24/7 to discuss her reporting.

More from CBS News

How low will mortgage rates drop after the Fed cuts rates? Experts weigh in

Don't pursue credit card debt forgiveness before considering these 4 things, experts say

4 expert-backed gold investing tips all seniors should know

More cinnamon recalled after tests find elevated levels of lead

How to Write a Research Proposal: (with Examples & Templates)

Table of Contents

Before conducting a study, a research proposal should be created that outlines researchers’ plans and methodology and is submitted to the concerned evaluating organization or person. Creating a research proposal is an important step to ensure that researchers are on track and are moving forward as intended. A research proposal can be defined as a detailed plan or blueprint for the proposed research that you intend to undertake. It provides readers with a snapshot of your project by describing what you will investigate, why it is needed, and how you will conduct the research.  

Your research proposal should aim to explain to the readers why your research is relevant and original, that you understand the context and current scenario in the field, have the appropriate resources to conduct the research, and that the research is feasible given the usual constraints.  

This article will describe in detail the purpose and typical structure of a research proposal , along with examples and templates to help you ace this step in your research journey.  

What is a Research Proposal ?  

A research proposal¹ ,²  can be defined as a formal report that describes your proposed research, its objectives, methodology, implications, and other important details. Research proposals are the framework of your research and are used to obtain approvals or grants to conduct the study from various committees or organizations. Consequently, research proposals should convince readers of your study’s credibility, accuracy, achievability, practicality, and reproducibility.   

With research proposals , researchers usually aim to persuade the readers, funding agencies, educational institutions, and supervisors to approve the proposal. To achieve this, the report should be well structured with the objectives written in clear, understandable language devoid of jargon. A well-organized research proposal conveys to the readers or evaluators that the writer has thought out the research plan meticulously and has the resources to ensure timely completion.  

Purpose of Research Proposals  

A research proposal is a sales pitch and therefore should be detailed enough to convince your readers, who could be supervisors, ethics committees, universities, etc., that what you’re proposing has merit and is feasible . Research proposals can help students discuss their dissertation with their faculty or fulfill course requirements and also help researchers obtain funding. A well-structured proposal instills confidence among readers about your ability to conduct and complete the study as proposed.  

Research proposals can be written for several reasons:³  

• To describe the importance of research in the specific topic  
• Address any potential challenges you may encounter  
• Showcase knowledge in the field and your ability to conduct a study  
• Apply for a role at a research institute  
• Convince a research supervisor or university that your research can satisfy the requirements of a degree program  
• Highlight the importance of your research to organizations that may sponsor your project  
• Identify implications of your project and how it can benefit the audience  

What Goes in a Research Proposal?    

Research proposals should aim to answer the three basic questions—what, why, and how.  

The What question should be answered by describing the specific subject being researched. It should typically include the objectives, the cohort details, and the location or setting.  

The Why question should be answered by describing the existing scenario of the subject, listing unanswered questions, identifying gaps in the existing research, and describing how your study can address these gaps, along with the implications and significance.  

The How question should be answered by describing the proposed research methodology, data analysis tools expected to be used, and other details to describe your proposed methodology.   

Research Proposal Example  

Here is a research proposal sample template (with examples) from the University of Rochester Medical Center. 4 The sections in all research proposals are essentially the same although different terminology and other specific sections may be used depending on the subject.  

Structure of a Research Proposal  

If you want to know how to make a research proposal impactful, include the following components:¹  

1. Introduction  

This section provides a background of the study, including the research topic, what is already known about it and the gaps, and the significance of the proposed research.  

2. Literature review  

This section contains descriptions of all the previous relevant studies pertaining to the research topic. Every study cited should be described in a few sentences, starting with the general studies to the more specific ones. This section builds on the understanding gained by readers in the Introduction section and supports it by citing relevant prior literature, indicating to readers that you have thoroughly researched your subject.  

3. Objectives  

Once the background and gaps in the research topic have been established, authors must now state the aims of the research clearly. Hypotheses should be mentioned here. This section further helps readers understand what your study’s specific goals are.  

4. Research design and methodology  

Here, authors should clearly describe the methods they intend to use to achieve their proposed objectives. Important components of this section include the population and sample size, data collection and analysis methods and duration, statistical analysis software, measures to avoid bias (randomization, blinding), etc.  

5. Ethical considerations  

This refers to the protection of participants’ rights, such as the right to privacy, right to confidentiality, etc. Researchers need to obtain informed consent and institutional review approval by the required authorities and mention this clearly for transparency.  

6. Budget/funding  

Researchers should prepare their budget and include all expected expenditures. An additional allowance for contingencies such as delays should also be factored in.  

7. Appendices  

This section typically includes information that supports the research proposal and may include informed consent forms, questionnaires, participant information, measurement tools, etc.  

8. Citations  

Important Tips for Writing a Research Proposal  

Writing a research proposal begins much before the actual task of writing. Planning the research proposal structure and content is an important stage, which if done efficiently, can help you seamlessly transition into the writing stage. 3,5  

The Planning Stage  

• Manage your time efficiently. Plan to have the draft version ready at least two weeks before your deadline and the final version at least two to three days before the deadline.
• What is the primary objective of your research?  
• Will your research address any existing gap?  
• What is the impact of your proposed research?  
• Do people outside your field find your research applicable in other areas?  
• If your research is unsuccessful, would there still be other useful research outcomes?  

  The Writing Stage  

• Create an outline with main section headings that are typically used.  
• Focus only on writing and getting your points across without worrying about the format of the research proposal , grammar, punctuation, etc. These can be fixed during the subsequent passes. Add details to each section heading you created in the beginning.   
• Ensure your sentences are concise and use plain language. A research proposal usually contains about 2,000 to 4,000 words or four to seven pages.  
• Don’t use too many technical terms and abbreviations assuming that the readers would know them. Define the abbreviations and technical terms.  
• Ensure that the entire content is readable. Avoid using long paragraphs because they affect the continuity in reading. Break them into shorter paragraphs and introduce some white space for readability.  
• Focus on only the major research issues and cite sources accordingly. Don’t include generic information or their sources in the literature review.  
• Proofread your final document to ensure there are no grammatical errors so readers can enjoy a seamless, uninterrupted read.  
• Use academic, scholarly language because it brings formality into a document.  
• Ensure that your title is created using the keywords in the document and is neither too long and specific nor too short and general.  
• Cite all sources appropriately to avoid plagiarism.  
• Make sure that you follow guidelines, if provided. This includes rules as simple as using a specific font or a hyphen or en dash between numerical ranges.  
• Ensure that you’ve answered all questions requested by the evaluating authority.  

Key Takeaways   

Here’s a summary of the main points about research proposals discussed in the previous sections:  

• A research proposal is a document that outlines the details of a proposed study and is created by researchers to submit to evaluators who could be research institutions, universities, faculty, etc.  
• Research proposals are usually about 2,000-4,000 words long, but this depends on the evaluating authority’s guidelines.  
• A good research proposal ensures that you’ve done your background research and assessed the feasibility of the research.  
• Research proposals have the following main sections—introduction, literature review, objectives, methodology, ethical considerations, and budget.  

Frequently Asked Questions  

Q1. How is a research proposal evaluated?  

A1. In general, most evaluators, including universities, broadly use the following criteria to evaluate research proposals . 6  

• Significance —Does the research address any important subject or issue, which may or may not be specific to the evaluator or university?  
• Content and design —Is the proposed methodology appropriate to answer the research question? Are the objectives clear and well aligned with the proposed methodology?  
• Sample size and selection —Is the target population or cohort size clearly mentioned? Is the sampling process used to select participants randomized, appropriate, and free of bias?  
• Timing —Are the proposed data collection dates mentioned clearly? Is the project feasible given the specified resources and timeline?  
• Data management and dissemination —Who will have access to the data? What is the plan for data analysis?  

Q2. What is the difference between the Introduction and Literature Review sections in a research proposal ?  

A2. The Introduction or Background section in a research proposal sets the context of the study by describing the current scenario of the subject and identifying the gaps and need for the research. A Literature Review, on the other hand, provides references to all prior relevant literature to help corroborate the gaps identified and the research need.  

Q3. How long should a research proposal be?  

A3. Research proposal lengths vary with the evaluating authority like universities or committees and also the subject. Here’s a table that lists the typical research proposal lengths for a few universities.  

Q4. What are the common mistakes to avoid in a research proposal ?  

A4. Here are a few common mistakes that you must avoid while writing a research proposal . 7  

• No clear objectives: Objectives should be clear, specific, and measurable for the easy understanding among readers.  
• Incomplete or unconvincing background research: Background research usually includes a review of the current scenario of the particular industry and also a review of the previous literature on the subject. This helps readers understand your reasons for undertaking this research because you identified gaps in the existing research.  
• Overlooking project feasibility: The project scope and estimates should be realistic considering the resources and time available.   
• Neglecting the impact and significance of the study: In a research proposal , readers and evaluators look for the implications or significance of your research and how it contributes to the existing research. This information should always be included.  
• Unstructured format of a research proposal : A well-structured document gives confidence to evaluators that you have read the guidelines carefully and are well organized in your approach, consequently affirming that you will be able to undertake the research as mentioned in your proposal.  
• Ineffective writing style: The language used should be formal and grammatically correct. If required, editors could be consulted, including AI-based tools such as Paperpal , to refine the research proposal structure and language.  

Thus, a research proposal is an essential document that can help you promote your research and secure funds and grants for conducting your research. Consequently, it should be well written in clear language and include all essential details to convince the evaluators of your ability to conduct the research as proposed.  

This article has described all the important components of a research proposal and has also provided tips to improve your writing style. We hope all these tips will help you write a well-structured research proposal to ensure receipt of grants or any other purpose.  

References  

• Sudheesh K, Duggappa DR, Nethra SS. How to write a research proposal? Indian J Anaesth. 2016;60(9):631-634. Accessed July 15, 2024. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5037942/  
• Writing research proposals. Harvard College Office of Undergraduate Research and Fellowships. Harvard University. Accessed July 14, 2024. https://uraf.harvard.edu/apply-opportunities/app-components/essays/research-proposals  
• What is a research proposal? Plus how to write one. Indeed website. Accessed July 17, 2024. https://www.indeed.com/career-advice/career-development/research-proposal  
• Research proposal template. University of Rochester Medical Center. Accessed July 16, 2024. https://www.urmc.rochester.edu/MediaLibraries/URMCMedia/pediatrics/research/documents/Research-proposal-Template.pdf  
• Tips for successful proposal writing. Johns Hopkins University. Accessed July 17, 2024. https://research.jhu.edu/wp-content/uploads/2018/09/Tips-for-Successful-Proposal-Writing.pdf  
• Formal review of research proposals. Cornell University. Accessed July 18, 2024. https://irp.dpb.cornell.edu/surveys/survey-assessment-review-group/research-proposals  
• 7 Mistakes you must avoid in your research proposal. Aveksana (via LinkedIn). Accessed July 17, 2024. https://www.linkedin.com/pulse/7-mistakes-you-must-avoid-your-research-proposal-aveksana-cmtwf/  

Paperpal is a comprehensive AI writing toolkit that helps students and researchers achieve 2x the writing in half the time. It leverages 21+ years of STM experience and insights from millions of research articles to provide in-depth academic writing, language editing, and submission readiness support to help you write better, faster.  

Get accurate academic translations, rewriting support, grammar checks, vocabulary suggestions, and generative AI assistance that delivers human precision at machine speed. Try for free or upgrade to Paperpal Prime starting at US$19 a month to access premium features, including consistency, plagiarism, and 30+ submission readiness checks to help you succeed.  

Experience the future of academic writing – Sign up to Paperpal and start writing for free!  

Related Reads:

• How to Paraphrase Research Papers Effectively
• How to Cite Social Media Sources in Academic Writing? 
• What is the Importance of a Concept Paper and How to Write It 

APA format: Basic Guide for Researchers

You may also like, how to choose a dissertation topic, how to write a phd research proposal, how to write an academic paragraph (step-by-step guide), five things authors need to know when using..., 7 best referencing tools and citation management software..., maintaining academic integrity with paperpal’s generative ai writing..., research funding basics: what should a grant proposal..., how to write an abstract in research papers..., how to write dissertation acknowledgements.

Persuasive Essay On Animal Testing

How would you feel if you were an animal ,that got randomly selected to get an experiment done on ? yeah, very scary huh ? Well in my research paper i am putting a view on that situation, and how the selected animals should be recognized and rewarded for their risky life being used. Animals are just as scared ,then humans are in any situation, so why not reward them for their disabilities. With that being said, I am willing to put a stop on the animal disability on their choice of being tested. Animals should never be experimented, nor tested on . The result of animal testing could harmfully lead to a major injury , infection , and even death. If an animal is denying the force to not be tested , than that means no. Just because …show more content…

In 2009 , it was proven that more than 300 cases were held due to animal cruelty because of animal testing. We don’t need anymore cases to be on animals due to them dying or close to dying because they were tested . Furthermore, it is still a unwanted act. Using mice and rats to test the safety of drugs in humans are only accurate 43% of the time, so why not just use humans to test it on since we are the ones who’s going to use it anyways. We shouldn’t make animals suffer because we want something. Think of animals as if they were yours, don’t hurt them. Let’s put yourself in the animals shoes, humans love chocolate, raisins, macadamia nuts , and grapes, but those things are toxic to dogs. If you experiment that on a dog first, then it might wouldn’t have even made it to humans because it’s so toxic to dogs. Instead of animal testing they should try human testing sounds a lot better. In the end the moral status of animals is questioned. Are they on the level of people, property, or pets.Most would not put animals in the same category as humans so giving them the same rights seem dumb to us. Which is not a legitimate reason to say it’s of and treat animals like non-living

More about Persuasive Essay On Animal Testing

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Jt Dis Relat Surg
• v.33(2); 2022 Aug

The dark side of the animal experiments

Elif akkaya.

1 Uskudar American Academy, High School, Istanbul, Türkiye

Harun Reşit Güngör

2 Department of Orthopedics and Traumatology, Pamukkale University Faculty of Medicine, Denizli, Türkiye

The use of animals in research has increasingly continued, although there are serious concerns about appropriate methodology, moral issues and translation to clinical practice. The aim of the present article is to review the insufficiency of statements in ethical approvals to obtain animal welfare in real life practices and to draw attention to the need for an additional evidence-based audit system. In many countries, local Animal Ethics Committees are established to ensure that animals are treated in accordance with 3Rs (Replace, Reduce, Refine) principles. Although the ethical approval certificate which is taken before the start of research is considered as the proof of animal welfare, footages released from all over the world reveal the maltreatment. However, due to the scientific resource provided by animal models, it does not seem possible to expect animal experiments to be terminated in the near future. Addition of previously suggested welfare section to methods of study or including the ethical approval certificate does not seem to be sufficient practices to guarantee the animal welfare, since they are not based on audited evidence. The welfare certificate, in which the welfare is supervised by independent auditors, would serve as a proof of both the wellbeing of subjects and consequently the scientific reliability of data. Application of review and publication priority for the animal researches which have the welfare certificate in addition to the ethical approval would encourage the researchers to obtain this certificate. The achievement of worldwide consensus about content, requirements, and application methods of the welfare certificate should be in the scope of scientists in the near future to reach the more humane and more qualified animal experiments.

Many animals have similar body compartments and physiology as humans; therefore, they are used in scientific research and studies. The use of animals in medical research has a long history dating to the anatomical studies of Aristotle on various animals. Although there are serious concerns about the appropriate methodology and moral issues in animal studies and the transfer of data to humans, the use of animals in experiments has increasingly continued to the present day.[ 1 ] Currently, millions of animals are sacrificed worldwide annually, and many are suffering in harmful conditions for medical research involving the development of several medicines, vaccines, or surgical techniques.[ 2 - 4 ]

The 3Rs (Replace, Reduce, Refine) are recommended as fundamental principles for the use of animal model.[ 5 ] Replacement involves the alternative models to animal experiments such as in vitro methods or computer modelling. Reduction means to propose the minimum number of animals required to achieve the purpose of the research. Refinement consists of many different applications to minimize the suffering, distress, and the potential pain exposure throughout the research.[ 5 , 6 ] Guidelines for research ethics stipulate the principles to be applied throughout the scientific study from planning to publication and recommend a set of ethical standards including the welfare of animal subjects. Directive 2010/63/EU of the European Parliament and of the Council on the protection of animals used for scientific purposes designated that the measures should be taken for animal welfare and that animal testing is replaced, when there are possible alternative methods.[ 7 ]

All over the world in many countries, local Animal Ethics Committees are established to ensure that animals are treated in accordance with 3Rs principles.[ 5 ] In Türkiye, institutional local Animal Ethical Committees approve, revise or reject each application, whereas ethical committees in some countries have only consultant role.[ 5 , 8 ] Researchers are expected to perform all experimental applications in accordance with 3Rs principles and to take necessary measures for animals to prevent suffering, distress, and pain.[ 3 , 9 ] Whereas replacement and reduction principles can be controlled and modified by the supervision of ethical committee before the approval of the experimental study, among the 3Rs principles, refinement is the most ambiguous one and the least monitored rule.[ 5 ] Several authors have suggested adding an animal welfare section to the methods part of publications to ensure that the refinement rule of 3Rs principles has been properly applied.[ 10 , 11 ] Although the certificate of ethical committee approval which is taken before the start of research is considered as the proof of animal subjects’ welfare based on ethical and scientific standards, many footages released from many experienced animal research centers all over the world reveal the maltreatment and abuse of animals.[ 12 - 15 ] Addition of previously suggested welfare section to the methods part of the study or claiming that the refinement principle of the 3Rs is complied with in the ethics committee document does not seem to be sufficient practices to guarantee the welfare of animals, since they are not based on audited evidence. Therefore, although ethical guidelines and approvals are useful and required tools to establish the ethical limits of experiments on animals, there is a clear need for a new measure to achieve the welfare of animal subjects.

Examples of animal cruelty in the international media reveal the insufficient application of 3Rs in real life for a variety of reasons and the urge need for an evidence-based audit system throughout the research in addition to the ethical approval. First, some laboratories continue to neglect their animals or ignore the negative effects of certain practices on animals unrelated to the scientific purpose of the experiment as long as they achieve their goals. One of the largest primate research centers in the United States (US) - University of California, Davis (UC Davis) has also been the target of animal rights activists over several mistreatments of primates.[ 13 ] In 2005, the facility was fined, when seven monkeys died from exposure to extreme heat. Additionally in 2016, UC Davis was investigated about a primate which broke its legs, while escaping through an unsafe door and another primate which was injured in a similar incident. UC Davis has typically used dyes to identify individual primates; however, in 2018, a few weeks old seven baby monkeys died due to the toxic allergic reaction caused by the dye accidentally transferred from their mothers, although the baby monkeys were not the subjects of that experiment.[ 13 ] Both repeated similar incidences and addition of new different maltreatments by years from the same research center, even if it is a specialized center for these researches, suggest that ethical approval could not be the guarantee of animal subjects’ welfare.

Another reason why additional measure besides ethical approval is needed to ensure the well-being of animal subjects is that some staff abuses or makes fun of animals in the laboratory. To illustrate, in a footage published by Cruelty Free International from an animal experimentation laboratory in Spain, Vivotecnia, a male monkey was seen as pinned down on the table by a staff member who was collecting blood from its leg, while a senior staff member was seen as drawing a face on the monkey’s genitals. Vivotecnia is a Madrid-based research center conducting animal experiments for the biopharmaceutical, cosmetic, chemical, tobacco, and food industries from all over the world.[ 14 ] In such experienced experimental centers, laboratory workers after a while may begin to ignore that experimental animals are actually living beings and begin to think of them as tools that lead to a goal.

Finally, most of the animals in experiments may suffer from unnecessary pain and stress beyond the aim of experiment. Exposure of animals to this unnecessary stress affects not only the welfare of the animals, but also the scientific reliability of collected data. However, laboratory staff do not care that animals are exposed to this unnecessary stress as long as they can collect the data. A footage published by Cruelty Free International and Soko Tierschutz from a toxicology laboratory in Germany revealed the maltreatments of animal subjects for toxicology and dose ranging experiments. The Laboratory of Pharmacology and Toxicology carries out toxicity tests for agrochemical, pharmaceutical, and industrial companies worldwide. As evident in the footage from toxicology tests, although it did not serve for the purpose of the trial, it was ignored that the monkeys - fixed by their necks while waiting their turn - were exposed to watching the applications of experiment on other monkeys and experienced intense stress.[ 15 ] Although animal toxicity tests include the application of the dose causing serious harm to animal subjects, in an attempt to predict what a safe dose for humans may be, the results are not actual predictors of safety and effectivity in humans. Yet, the reactions to a particular substance for all species and humans might be quite different.[ 16 , 17 ] Moreover, collected data becomes unreliable due to the physiological changes in animal subjects caused by the severe stress exposure beyond the purpose of experiment.[ 18 ] Consequently, obtaining unreliable data due to unnecessary stress exposure of animals in addition to the scientific limitations of toxicology testing methods for predicting the safety of human use actually refers that many animals are wasted for nothing. However, if toxicology experiments are required, the endpoint should be clearly defined according to the purpose of the experiment and necessary measures should be taken to ensure both animal welfare and data reliability by preventing animal subjects from suffering pain, distress and abuse beyond the purpose of research.

In addition to these examples of abuse in animal experiments, a topic that has been discussed recently is that although thousands of animals are spent on scientific experiments, the majority of these studies are not published[ 3 , 19 - 23 ] or the impact of the published ones on scientific progress is controversial due to the limited transfer of animal experimental data to humans.[ 3 , 17 , 19 , 20 ] Öztürk and Ersan[ 3 ] reported that more than 40% of animal experiments that were represented at national orthopedic congress in Türkiye over a nine-year period were never published, and 38% of those that were published never cited or were cited only once. They found that 4,440 animals were euthanized for no obvious scientific gain in unpublished studies.[ 3 ] Unpublished animal studies result in waste of animal life.[ 3 , 19 - 23 ] Öztürk and Ersan[ 3 ] suggested that publishing even animal studies that did not find significant differences would be helpful to avoid duplication of the same study. There are guidelines such as the Animal Research: Reporting of In vivo Experiments (ARRIVE) guidelines to reduce the unnecessary animal experiments and to overcome the inadequate methodology in animal studies.[ 24 ] Researchers are expected to comply with the ARRIVE criteria and make the necessary effort to publish the results of animal studies and, therefore, the lives of animals sacrificed in experimental studies are not wasted for nothing.[ 3 , 19 , 20 ] van der Worp et al.[ 25 ] suggested that the only way to prevent the unreported data in animal studies was to have a central registry system for animal studies similarly to clinical trial registry systems.

Since animals are known to have similar ability to feel pain and enjoy life as humans, it would be morally unacceptable to treat animals only as ‘tools’ to advance the knowledge.[ 26 ] Ignoring the fact that animal subjects are also living beings during experiments may expose them to unnecessary suffering and stress beyond the scientific purpose of the experiment. However, due to the scientific resource provided by animal models, it does not seem possible to expect animal experiments to be terminated all over the world in the near future. Moreover, the US Food and Drug Administration (FDA) usually asks two or more animal tests before approval of human trials.[ 20 ] In this case, the recurrence of many negative examples that have been reported previously should be prevented by taking more strict measures in centers of animal experiments. Although requirement of ethics committee approval for publication acceptance is applied all over the world, ethics committee approval does not ensure the animal welfare and data reliability.[ 12 ] When an experimental study is carried out with the approval of the ethics committee, researchers are expected to comply with the 3Rs principles. Since the refinement principle is the most uncertain among these 3Rs principles and can be achieved by various methods and there is no consensus about it, the ethics committee approval of the study, unfortunately, cannot provide definitive proof of animal welfare.

In conclusion, since animals do not have the chance to defend their rights such as humans, there is a need for a supervisor mechanism independent of the researcher to supervise and report whether the welfare principle is actually met in experimental animal studies in the reallife practice. The welfare certificate, in which the welfare of the subjects is supervised during the experiment, would serve as a proof of both the well-being of the subjects and the consequently scientific reliability of the data.[ 27 ] In this context, the content of the welfare certificate includes the criteria that the researcher is routinely expected to comply with (humane endpoints, appropriate skills and training of researchers to minimize pain and distress experienced by animals, taking measures to reduce pain and distress, improved handling of animals, appropriate living conditions, etc.), but only the researcher's statement is not sufficient, and it should be also audited that the necessary conditions are met with the application of the evidence-based control mechanism. Audit of adherence to welfare certificate criteria could be checked with regular video records of applications and treatments which were sent to welfare audit organization, and additional spontaneous control visits applied by the welfare experts. Application of a review and publication priority for the animal experiments which have the welfare certificate in addition to the ethical approval certificate would encourage the researchers to achieve this welfare document. In this respect, editors and journals would have an important role and sanction power for the improvement of animal welfare in experiments.[ 28 ] The achievement of worldwide consensus about the content, the requirements, and the application methods of the welfare certificate should be in the scope of scientists in the near future to reach the more humane and more qualified animal experiments.

Conflict of Interest: The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Author Contributions: Devised the project, the main conceptual ideas and proof outline: E.E.; Interpreted the data, revised critically for the intellectual content: H.R.G.; Both authors approved the final version to be published.

Financial Disclosure: The authors received no financial support for the research and/or authorship of this article.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Published: 25 July 2024

Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle

• Leonardo C. Caserta   ORCID: orcid.org/0000-0003-1643-8560 1   na1 ,
• Elisha A. Frye   ORCID: orcid.org/0000-0002-7784-6771 1   na1 ,
• Salman L. Butt 1   na1 ,
• Melissa Laverack 1 ,
• Mohammed Nooruzzaman   ORCID: orcid.org/0000-0002-9358-1494 1 ,
• Lina M. Covaleda 1 ,
• Alexis C. Thompson   ORCID: orcid.org/0000-0003-4405-5313 2 ,
• Melanie Prarat Koscielny 4 ,
• Brittany Cronk   ORCID: orcid.org/0000-0001-7239-4262 1 ,
• Ashley Johnson 4 ,
• Katie Kleinhenz 2 ,
• Erin E. Edwards   ORCID: orcid.org/0000-0001-9447-8395 3 ,
• Gabriel Gomez 3 ,
• Gavin Hitchener 1 ,
• Mathias Martins   ORCID: orcid.org/0000-0002-8290-5756 3 ,
• Darrell R. Kapczynski 5 ,
• David L. Suarez 5 ,
• Ellen Ruth Alexander Morris   ORCID: orcid.org/0000-0002-7957-4642 3 ,
• Terry Hensley 3 ,
• John S. Beeby 1 ,
• Manigandan Lejeune 1 ,
• Amy K. Swinford 3 ,
• François Elvinger 1 ,
• Kiril M. Dimitrov   ORCID: orcid.org/0000-0002-5525-4492 3 &
• Diego G. Diel   ORCID: orcid.org/0000-0003-3237-8940 1  

Nature ( 2024 ) Cite this article

17k Accesses

1300 Altmetric

Metrics details

We are providing an unedited version of this manuscript to give early access to its findings. Before final publication, the manuscript will undergo further editing. Please note there may be errors present which affect the content, and all legal disclaimers apply.

• Influenza virus
• Molecular evolution

Highly pathogenic avian influenza (HPAI) H5N1 clade 2.3.4.4b virus has caused the death of millions of domestic birds and thousands of wild birds in the U.S. since January, 2022 1–4 Throughout this outbreak, spillovers to mammals have been frequently documented 5–12 . We report spillover of HPAI H5N1 virus in dairy cattle herds across several states in the U.S. The affected cows displayed clinical signs encompassing decreased feed intake, altered fecal consistency, respiratory distress, and decreased milk production with abnormal milk. Infectious virus and viral RNA were consistently detected in milk from affected cows. Viral distribution in tissues via immunohistochemistry and in situ hybridization revealed a distinct tropism of the virus for the epithelial cells lining the alveoli of the mammary gland in cows. Whole viral genome sequences recovered from dairy cows, birds, domestic cats, and a raccoon from affected farms indicated multidirectional interspecies transmissions. Epidemiologic and genomic data revealed efficient cow-to-cow transmission after apparently healthy cows from an affected farm were transported to a premise in a different state. These results demonstrate the transmission of HPAI H5N1 clade 2.3.4.4b virus at a non-traditional interface underscoring the ability of the virus to cross species barriers.

You have full access to this article via your institution.

Similar content being viewed by others

Pathogenicity and transmissibility of bovine H5N1 influenza virus

Risk assessment of a highly pathogenic H5N1 influenza virus from mink

Persistence of viral RNA in North American elk experimentally infected with an ancestral strain of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)

Author information.

These authors contributed equally: Leonardo C. Caserta, Elisha A. Frye, Salman L. Butt

Authors and Affiliations

Department of Population Medicine and Diagnostic Sciences, Animal Health Diagnostic Center, College of Veterinary Medicine, Cornell University, Ithaca, NY, USA

Leonardo C. Caserta, Elisha A. Frye, Salman L. Butt, Melissa Laverack, Mohammed Nooruzzaman, Lina M. Covaleda, Brittany Cronk, Gavin Hitchener, John S. Beeby, Manigandan Lejeune, François Elvinger & Diego G. Diel

Texas A&M Veterinary Medical Diagnostic Laboratory, Canyon, TX, USA

Alexis C. Thompson & Katie Kleinhenz

Texas A&M Veterinary Medical Diagnostic Laboratory, College Station, TX, USA

Erin E. Edwards, Gabriel Gomez, Mathias Martins, Ellen Ruth Alexander Morris, Terry Hensley, Amy K. Swinford & Kiril M. Dimitrov

Ohio Animal Disease and Diagnostic Laboratory, Ohio Department of Agriculture, Reynoldsburg, OH, USA

Melanie Prarat Koscielny & Ashley Johnson

Southeast Poultry Research Laboratory, U.S. National Poultry Research Center, Agricultural Research Service, United States Department of Agriculture, Athens, GA, USA

Darrell R. Kapczynski & David L. Suarez

You can also search for this author in PubMed   Google Scholar

Corresponding authors

Correspondence to Kiril M. Dimitrov or Diego G. Diel .

Supplementary information

Supplementary discussion.

This file contains Supplementary discussion that complements the discussion section of the main text, and Supplementary References.

Reporting Summary

Supplementary table 1.

RT-PCR data in CSV format showing testing results for Influenza A Matrix and H5-specific RT-PCR on all samples tested in the present study.

Supplementary Table 2

RT-PCR data in CSV format showing testing results for Influenza A Matrix and H5-specific RT-PCR on serial samples collected on days 3, 16 and 31 post-clinical diagnosis (pcd).

Supplementary Table 3

Mutational analysis in CVS format showing all amino acid mutations identified in PB2, PB1, PA, HA, NA, M (M1 and M2), NS (NS1 and NS2) gene segments of the highly pathogenic influenza A H5N1 virus sequences obtained in the present study. Predicted phenotype changes as reported through FluServer (GISAID) are also included.

Supplementary Table 4

List of sequences including GISAID accession numbers in CVS format used in the phylogenomic and mutational analyses conducted in the present study.

Supplementary Table 5

List of highly pathogenic influenza A H5N1 virus sequences obtained in the present study including GISAID accession numbers in CVS format and sample metadata (H5 clade classification, genotype, collection date, state, host, sample type and premise).

Supplementary Table 6

Mutational analysis in CVS format showing amino acid mutations that accumulated over time and identified in PB2, PB1, PA, HA, NA, NP and NS1 genome segments of highly pathogenic influenza H5N1 virus and their frequency among the available H5N1 sequences derived from cattle.

Peer Review File

Rights and permissions.

Reprints and permissions

About this article

Cite this article.

Caserta, L.C., Frye, E.A., Butt, S.L. et al. Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle. Nature (2024). https://doi.org/10.1038/s41586-024-07849-4

Download citation

Received : 22 May 2024

Accepted : 18 July 2024

Published : 25 July 2024

DOI : https://doi.org/10.1038/s41586-024-07849-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Suggestions or feedback?

MIT News | Massachusetts Institute of Technology

• Machine learning
• Sustainability
• Black holes
• Classes and programs

Departments

• Aeronautics and Astronautics
• Brain and Cognitive Sciences
• Architecture
• Political Science
• Mechanical Engineering

Centers, Labs, & Programs

• Abdul Latif Jameel Poverty Action Lab (J-PAL)
• Picower Institute for Learning and Memory
• Lincoln Laboratory
• School of Architecture + Planning
• School of Engineering
• School of Humanities, Arts, and Social Sciences
• Sloan School of Management
• School of Science
• MIT Schwarzman College of Computing

Large language models don’t behave like people, even though we may expect them to

Press contact :, media download.

*Terms of Use:

Images for download on the MIT News office website are made available to non-commercial entities, press and the general public under a Creative Commons Attribution Non-Commercial No Derivatives license . You may not alter the images provided, other than to crop them to size. A credit line must be used when reproducing images; if one is not provided below, credit the images to "MIT."

Previous image Next image

One thing that makes large language models (LLMs) so powerful is the diversity of tasks to which they can be applied. The same machine-learning model that can help a graduate student draft an email could also aid a clinician in diagnosing cancer.

However, the wide applicability of these models also makes them challenging to evaluate in a systematic way. It would be impossible to create a benchmark dataset to test a model on every type of question it can be asked.

In a new paper , MIT researchers took a different approach. They argue that, because humans decide when to deploy large language models, evaluating a model requires an understanding of how people form beliefs about its capabilities.

For example, the graduate student must decide whether the model could be helpful in drafting a particular email, and the clinician must determine which cases would be best to consult the model on.

Building off this idea, the researchers created a framework to evaluate an LLM based on its alignment with a human’s beliefs about how it will perform on a certain task.

They introduce a human generalization function — a model of how people update their beliefs about an LLM’s capabilities after interacting with it. Then, they evaluate how aligned LLMs are with this human generalization function.

Their results indicate that when models are misaligned with the human generalization function, a user could be overconfident or underconfident about where to deploy it, which might cause the model to fail unexpectedly. Furthermore, due to this misalignment, more capable models tend to perform worse than smaller models in high-stakes situations.

“These tools are exciting because they are general-purpose, but because they are general-purpose, they will be collaborating with people, so we have to take the human in the loop into account,” says study co-author Ashesh Rambachan, assistant professor of economics and a principal investigator in the Laboratory for Information and Decision Systems (LIDS).

Rambachan is joined on the paper by lead author Keyon Vafa, a postdoc at Harvard University; and Sendhil Mullainathan, an MIT professor in the departments of Electrical Engineering and Computer Science and of Economics, and a member of LIDS. The research will be presented at the International Conference on Machine Learning.

Human generalization

As we interact with other people, we form beliefs about what we think they do and do not know. For instance, if your friend is finicky about correcting people’s grammar, you might generalize and think they would also excel at sentence construction, even though you’ve never asked them questions about sentence construction.

“Language models often seem so human. We wanted to illustrate that this force of human generalization is also present in how people form beliefs about language models,” Rambachan says.

As a starting point, the researchers formally defined the human generalization function, which involves asking questions, observing how a person or LLM responds, and then making inferences about how that person or model would respond to related questions.

If someone sees that an LLM can correctly answer questions about matrix inversion, they might also assume it can ace questions about simple arithmetic. A model that is misaligned with this function — one that doesn’t perform well on questions a human expects it to answer correctly — could fail when deployed.

With that formal definition in hand, the researchers designed a survey to measure how people generalize when they interact with LLMs and other people.

They showed survey participants questions that a person or LLM got right or wrong and then asked if they thought that person or LLM would answer a related question correctly. Through the survey, they generated a dataset of nearly 19,000 examples of how humans generalize about LLM performance across 79 diverse tasks.

Measuring misalignment

They found that participants did quite well when asked whether a human who got one question right would answer a related question right, but they were much worse at generalizing about the performance of LLMs.

“Human generalization gets applied to language models, but that breaks down because these language models don’t actually show patterns of expertise like people would,” Rambachan says.

People were also more likely to update their beliefs about an LLM when it answered questions incorrectly than when it got questions right. They also tended to believe that LLM performance on simple questions would have little bearing on its performance on more complex questions.

In situations where people put more weight on incorrect responses, simpler models outperformed very large models like GPT-4.

“Language models that get better can almost trick people into thinking they will perform well on related questions when, in actuality, they don’t,” he says.

One possible explanation for why humans are worse at generalizing for LLMs could come from their novelty — people have far less experience interacting with LLMs than with other people.

“Moving forward, it is possible that we may get better just by virtue of interacting with language models more,” he says.

To this end, the researchers want to conduct additional studies of how people’s beliefs about LLMs evolve over time as they interact with a model. They also want to explore how human generalization could be incorporated into the development of LLMs.

“When we are training these algorithms in the first place, or trying to update them with human feedback, we need to account for the human generalization function in how we think about measuring performance,” he says.

In the meanwhile, the researchers hope their dataset could be used a benchmark to compare how LLMs perform related to the human generalization function, which could help improve the performance of models deployed in real-world situations.

“To me, the contribution of the paper is twofold. The first is practical: The paper uncovers a critical issue with deploying LLMs for general consumer use. If people don’t have the right understanding of when LLMs will be accurate and when they will fail, then they will be more likely to see mistakes and perhaps be discouraged from further use. This highlights the issue of aligning the models with people's understanding of generalization,” says Alex Imas, professor of behavioral science and economics at the University of Chicago’s Booth School of Business, who was not involved with this work. “The second contribution is more fundamental: The lack of generalization to expected problems and domains helps in getting a better picture of what the models are doing when they get a problem ‘correct.’ It provides a test of whether LLMs ‘understand’ the problem they are solving.”

This research was funded, in part, by the Harvard Data Science Initiative and the Center for Applied AI at the University of Chicago Booth School of Business.

Share this news article on:

Related links.

• Ashesh Rambachan
• Laboratory for Information and Decision Systems
• Department of Electrical Engineering and Computer Science
• Department of Economics

Related Topics

• Computer science and technology
• Artificial intelligence
• Human-computer interaction
• Laboratory for Information and Decision Systems (LIDS)
• School of Humanities Arts and Social Sciences

Related Articles

Reasoning skills of large language models are often overestimated

Technique improves the reasoning capabilities of large language models

Natural language boosts LLM performance in coding, planning, and robotics

Multi-AI collaboration helps reasoning and factual accuracy in large language models

Previous item Next item

More MIT News

Researchers return to Arctic to test integrated sensor nodes

Read full story →

Precision home robots learn with real-to-sim-to-real

Method prevents an AI model from being overconfident about wrong answers

Helping Olympic athletes optimize their performance, one stride at a time

New method enables fast, accurate estimates of cardiovascular state to inform blood pressure management

Across the pond to scale new heights

• More news on MIT News homepage →

Massachusetts Institute of Technology 77 Massachusetts Avenue, Cambridge, MA, USA

• Map (opens in new window)
• Events (opens in new window)
• People (opens in new window)
• Careers (opens in new window)
• Accessibility
• Social Media Hub
• MIT on Facebook
• MIT on YouTube
• MIT on Instagram