field experiment altruistic

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
• Open access
• Published: 01 November 2016

Altruistic punishment does not increase with the severity of norm violations in the field

• Loukas Balafoutas 1 ,
• Nikos Nikiforakis 2 &
• Bettina Rockenbach   ORCID: orcid.org/0000-0003-2624-1964 3  

Nature Communications volume  7 , Article number:  13327 ( 2016 ) Cite this article

6531 Accesses

46 Citations

164 Altmetric

Metrics details

• Human behaviour
• Social evolution

The degree of human cooperation among strangers is a major evolutionary puzzle. A prominent explanation is that cooperation is maintained because many individuals have a predisposition to punish those violating group-beneficial norms. A critical condition for cooperation to evolve in evolutionary models is that punishment increases with the severity of the violation. Here we present evidence from a field experiment with real-life interactions that, unlike in lab experiments, altruistic punishment does not increase with the severity of the violation, regardless of whether it is direct (confronting a violator) or indirect (withholding help). We also document growing concerns for counter-punishment as the severity of the violation increases, indicating that the marginal cost of direct punishment increases with the severity of violations. The evidence suggests that altruistic punishment may not provide appropriate incentives to deter large violations. Our findings thus offer a rationale for the emergence of formal institutions for promoting large-scale cooperation among strangers.

Similar content being viewed by others

People prefer coordinated punishment in cooperative interactions

Super-additive cooperation

Reputation effects drive the joint evolution of cooperation and social rewarding

Introduction.

The degree of human cooperation among strangers in large societies is a major evolutionary puzzle 1 , 2 . One of the most prominent explanations is that cooperation is maintained because many individuals have a predisposition to punish those violating group-beneficial norms, even if doing so is costly 1 , 2 , 3 . A critical condition for cooperation to be sustained in evolutionary models is that, as in formal institutions charged with maintaining social order 4 , altruistic punishment ‘fits the crime’, that is, it increases with the severity of the violation 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 . Although this condition appears to be satisfied in lab experiments 13 , 14 , 15 , 16 , 17 , 18 , the role of altruistic punishment in maintaining cooperation in daily life cannot be established without field data 2 , 3 . Controlled experiments have recently been carried out in natural field settings documenting the predisposition to punish norm violators 19 , 20 , but there is no evidence to date on whether altruistic punishment is responsive to the severity of norm violations.

Here we present findings from the first field experiment investigating whether punishment ‘fits the crime’ in real-life interactions. The experiment was conducted at the two largest train stations in Cologne, Germany, to ensure that interactions were most likely one-shot. A team of confederates violated the social norm of not littering in public areas under different experimental conditions which differed only in the severity of the norm violation and the means available for punishing. The ‘small violation’ features a confederate (‘Violator’) noticeably throwing an empty coffee cup on the train platform close to a passenger, while in the case of the ‘large violation’ the item thrown on the platform was a large paper bag, including among others the same empty coffee cup ( Fig. 1 ). A confederate recorded the reaction of a passenger who was standing alone close to the violation and who observed the violation (‘Observer’) (for more details see ‘Notes on the experimental procedure’ in the Methods). The main variables of interest are whether the Observer punished the Violator, for example, by asking her to pick up the litter or reprimanding her for her action, as well as the intensity of the punishment in cases when it occurred. The intensity of punishment was classified by independent coders as being low, medium, or high based on the exact transcripts collected by our team of research assistants (see ‘Intensity of direct punishment’ in the Methods).

Panel ( a ) shows the small violation: littering by throwing a small coffee cup. Panel ( b ) shows the large violation: littering by throwing a lunch bag with several items, including a small coffee cup.

Punishment of norm violators can occur directly by means of confronting the violator as in the two treatments described above. Direct punishment is typically costly because it requires time and effort to enact, and the punisher bears the risk of retaliation 21 , 22 , 23 , 24 , which is why such punishment has been termed ‘altruistic’ 13 . However, punishment can also occur indirectly by means of withholding help 20 . Although helping requires effort, indirect punishment can also be costly, for instance when there are norms for helping others or when those withholding help suffer a reputational cost 1 . To investigate the possibility that indirect punishment responds to the severity of the norm violation, we implemented for each violation severity two treatments. In ‘Small Help’ and ‘Large Help’ the Violator, a few seconds after having thrown the litter, reached inside her bag and a pack of books accidentally fell out. This allowed the Observer to offer help to recover one or more books. Hence, in these two treatments, the Observer had the option to punish the Violator directly, punish indirectly by withholding help, punish both ways or not punish at all. If helping rates are smaller in the `Large Help' treatment than in the `Small Help', then we will have evidence that indirect punishment increases with the severity of the violation. The two treatments `Small No Help' and `Large No Help' are identical in the violation act, but lack the book-falling act.

Table 1 summarizes our experimental treatments. In each of the four treatments we gathered at least 100 observations. The sample size was chosen such that if the propensity to punish large violations relative to small violations in the field is similar to that observed in lab experiments 13 , 14 , we would detect the difference in punishment rates at the 5% level of significance with a very high probability (see ‘Power calculations and sample size’ in the Methods).

Additionally, we administered two surveys in the same location, but with different people ( Supplementary Methods and Supplementary Note 1 ). In the first survey—conducted parallel to the experiment—respondents were presented with one of the two violations (small or large), and asked how they would feel about it, how they would react to it (for example, punish, withhold help), as well as the reasons for their reactions ( N =510). In the second survey—administered after the experiment—respondents were presented with both violations and asked to make direct comparisons between the two 25 ( N =324). These surveys allow establishing that our treatment manipulation was successful and help us better understand individual motives in the experiment.

The experimental evidence suggests that a key condition in all models seeking to explain the evolution of cooperation by means of altruistic punishment is not met in the field: altruistic punishment does not increase with the severity of the norm violation. This finding holds both for direct punishment (confronting a violator) and indirect punishment (withholding help). We furthermore document growing concerns for counter-punishment as the severity of the violation increases, indicating that the marginal cost of direct punishment increases with the severity of the violation. Our findings suggest that altruistic punishment may not provide appropriate incentives to deter large violations.

Treatment manipulation

We use survey responses to evaluate whether passengers consider overall the large violation to be more severe than the small violation, whether the large violation triggers stronger negative emotions, and whether respondents believe the large violation should be more strongly reprimanded. In the first survey, when asked independently how bothered they would be by each violation, respondents were significantly more bothered by the large than the small violation (Mann–Whitney U, z =−3.059, P =0.002, N =510). In the second survey, when asked directly to compare the two violations, passengers were significantly more bothered by the large violation (Wilcoxon sign-rank, z= 9.39 , P <0.001, N =324) and judged it as being significantly more severe (Wilcoxon sign-rank, z= 6.621 , P <0.001, Wilcoxon sign-rank, N =324). This finding is important as negative emotions have been shown to be the proximate mechanism for altruistic punishment 13 . Finally, when asked whether small and large violators should be equally reprimanded or whether one of them should be more strongly reprimanded, respondents in the second survey stated on aggregate that large violators should be more strongly reprimanded (Wilcoxon sign-rank, z =5.06, P <0.001, N =324). Hence, our treatment manipulation succeeded in creating two conditions that differ in the severity of the norm violation as perceived by the population under consideration, the extent of negative emotions they trigger, and the extent people believe the violators should be reprimanded. If punishment ‘fits the crime’ in the field, we expect to observe more punishment for the large violation than for the small violation.

Direct punishment

The rates of direct punishment across treatments can be seen in Fig. 2 . Strikingly, despite the fact that passengers report significantly stronger negative emotions towards the large than the small violation and believe large violators should be more strongly reprimanded, direct punishment rates are very similar for the large and small violation, and never significantly different from each other. This pattern is true for the aggregate of the `No Help' and the `Help' treatments (14.85 versus 14.36%; χ 2 (1)=0.020, P =0.89, N =404; see Fig. 2a ) as well as for the treatments separately (No Help: 18.63 versus 21.00%; χ 2 (1)=0.179 , P =0.67, N =202; Help: 11.00 versus 7.84%; χ 2 (1)=0.591, P =0.44, N =202; see Fig. 2b ). Moreover, the intensity of direct punishment, that is, the way the norm violator is reprimanded, is not significantly different across violations ( χ 2 (2)=1.024 , P =0.60, N =59; see Fig. 3 and ‘Intensity of direct punishment’ in the Methods). Thus, we provide clear evidence that in the field direct punishment does not increase with the severity of the norm violation.

Panel ( a ) compares the rate of direct punishment across treatments with small and large violations. The difference in punishment rates for small and large violations is not statistically significant ( χ 2 (1)=0.020, P =0.89, N =404). Panel ( b ) presents the rate of direct punishment by treatment. The difference is statistically significant neither in the treatments without helping opportunities ( χ 2 (1)=0.179 , P =0.67, N =202) nor the treatments with helping opportunities ( χ 2 (1)=0.591, P =0.44, N =202). Direct punishment rates are significantly lower in the treatments with helping opportunities than in those without ( χ 2 (1)=8.753 , P =0.003, N =404). The bars indicate 95% confidence intervals.

The figure shows the incidences of low, medium and high intensity of direct punishment for small and large violations. The difference in the intensity of punishment for small and large violations is not statistically significant ( χ 2 (2)=1.024, P =0.60, N =59). The coding is explained in ‘Power calculations and sample size’ in the Methods.

Indirect punishment

We first note that direct punishment rates are lower in the treatments with helping opportunities (19.80% in `No Help' versus 9.41% in `Help'; χ 2 (1)=8.753 , P =0.003, N =404), indicating that the possibility of indirect punishment by means of withholding help crowds out the willingness to punish violators directly, as in (20). In the absence of a violation, the helping rate in (20)—measured in the same location and under identical conditions—was 42.9%. This rate is significantly higher than that in `Small Help' ( χ 2 (1)=20.162, P <0.001, N =177), and in `Large Help' ( χ 2 (1)=22.539 , P <0.001, N =179) indicating the use of indirect punishment in our experiment. Importantly, helping rates for large and small violators do not differ (13.00% in `Small Help' and 11.76% in `Large Help'; χ 2 (1)=0.071, P =0.790, N =202) suggesting that indirect punishment also does not ‘fit the crime’ in our experiment.

Lab evidence shows that the total amount individuals are willing to invest in punishing strangers increases with the severity of the violation 13 , 14 , 15 , 16 , 17 , 18 . The proximate mechanism for punishment is negative emotions triggered by the violation 13 . Since we observe a similar increase in magnitude in negative emotions in our experiment (see ‘Negative emotions in response to norm violations’ in the Methods), we should expect more punishment towards the large violators. Yet, we have seen that altruistic punishment does not increase with the severity of the violation in the field. Punishment rates do not differ, neither for direct nor for indirect punishment. Thus, there must be a fundamental difference between punishing in existing lab experiments and in our field experiment.

With regards to direct punishment, most respondents in our first survey who stated that they would not punish despite being bothered by the violation, stated that this was out of fear for being counter-punished by the violator (given by a total of 60% of respondents in that group). Counter-punishment raises the cost to punishers for enforcing cooperation 22 . Importantly, the fear of counter-punishment increases with the severity of the norm violation. As seen in Fig. 4 , 53.7% of respondents said they would not punish small violations for fear of counter-punishment, compared with 67% of respondents in the case of large violations (χ 2 (1)=3.884, P =0.049, N =211). Thus, unlike in most lab experiments where the marginal cost of punishment is either constant 13 , 16 , 17 , 18 or decreasing 14 , 15 , in the field the (perceived) marginal cost of punishment appears to increase with the severity of the violation. This evidence may reflect a belief that more severe violations convey information about the social orientation of the violator, their general disregard for social attitudes and therefore the likelihood to engage in other types of antisocial behaviour such as counter-punishing. For models of cooperation, this finding implies an increasing evolutionary disadvantage for punishers 26 , making direct punishment less likely to prevent large violations.

The figure shows the proportion of respondents who stated fear of counter-punishment as the reason why they would not apply direct punishment for each violation (Survey 1, question 4). The proportion is significantly higher for large violations ( χ 2 (1)=3.884, P =0.049, N =211). The bars indicate 95% confidence intervals.

How can we explain the fact that indirect punishment is also unresponsive to the severity of violations in our experiment? Unlike direct punishment, it seems unlikely that indirect punishment triggers fears of counter-punishment. Indeed, examining the open-ended responses to our first survey about why people would help violators or not, there was not a single case a respondent stated s/he would help (that is, not punish indirectly) out of fear of retaliation. The responses suggest respondents generally fall into one of two categories: those who withhold help in order to punish violators no matter how small the violation is, and those who would help both violators because ‘that’s the right thing to do’. To investigate this conjecture, in the second survey, respondents presented with the two violations were asked whether they would be more likely to help the small violator, the large violator, neither of the two, or both. Of the respondents who stated they would help at least one of the violators, 97.3% said they would be equally likely to help small and large violators. Of the respondents who said they would withhold help from at least one of the violators, 90.3% stated they would withhold help from both violators. Even if we limit our sample to respondents explicitly stating they would be more bothered by the large violation, 96.4% would treat the two violators equally: 74.6% said they would withhold help from both, and 21.8% said they would help both. This finding suggests that helping rates are not sensitive to the severity of the violation because, even though large violations are considered to be worse, small violations are sufficient to justify withholding help.

In summary, we present the first evidence to suggest that a key condition in all models seeking to explain the evolution of cooperation by means of altruistic punishment is not met in the field: altruistic punishment does not increase with the severity of the violation. This evidence indicates that altruistic punishment may not provide appropriate incentives for strangers to cooperate. This finding is the more remarkable as we studied a population characterized by strong norms of civic cooperation for which punishment has been shown to ‘fit the crime’ in the lab 27 . Our results thus provide a rationale for the evolution of mechanisms that reduce the threat of counter-punishment and, generally, provide appropriate incentives for cooperation such as pooled punishment 28 , 29 , 30 and formal institutions charged with maintaining social order among strangers. In addition, our findings indicate that counter-punishment should not be ignored in theoretical or experimental analyses. Although a few pioneering studies have explored the related concept of anti-social punishment 31 , 32 , 33 , more theoretical work is needed in order to understand how the willingness to retaliate punishment may have evolved 34 and its implications for the evolution of cooperation.

Notes on the experimental procedure

All aspects of the study, including ethical acceptability, were reviewed by the Vice-Rectorate for Research at the University of Innsbruck and permission was granted to conduct the experiment. The Deutsche Bahn gave written consent to running the experiment, which took place in May 2015 on various platforms in the two large (long-distance) train stations in Cologne, Germany. The data collection occurred between 9 am and 3 pm. The data were collected in five groups of two confederates each (one actor and one supervisor). All actors were female. The actors performed the violations by throwing an item on the platform so that a nearby observer saw it. The violators threw the item in an obviously intentional, but not provocative manner. Approximately 10–15 s later, in treatments `Small Help' and `Large Help', the actor ‘accidentally’ dropped some books in front of the selected observer (while trying to get something from the bottom of her shoulder bag).

Acts were performed only with single, standing observers to ensure there was no second-order public good problem with respect to punishing the violator and that helping was costly for individuals who had to bend to pick up the books. Observers were randomly assigned into treatments. Kruskal-Wallis tests using the observer’s gender, height, age, and the confederates, suggest the randomization was successful (gender: P =0.95; height: P =0.95; age: P =0.14; confederates: P =0.95). The P -values are from two-tailed tests, like all P -values reported in the paper.

Supervisors were standing at a distance of about 3–5 metres away from the scene of the violation and were instructed to remain passive during the entire interaction, and avoid any interference. Supervisors recorded only acts in which the observer did not leave the scene before the scripted act was complete and no other passenger approached. Observing the violation and then leaving the scene was coded as no punishment. Supervisors also ensured the acts were performed according to the script, and that observers witnessed the behaviours described in Acts 1 and 2 ( Table 1 ). Supervisors recorded whether the observer helped to pick up the books (in treatments Small Help and Large Help), whether he or she applied direct punishment against the violator (in all treatments), and the exact form that direct punishment took. Whenever an observer picked up at least one book, his or her action was recorded as help. Whenever an observer explicitly asked the violator to pick up the cup, or expressed disapproval of the norm violation his or her action was recorded as direct punishment.

Supervisors also recorded the following information: time of day when the observation was collected, an estimate of the observer’s age, activity of observer while waiting (for example, eating, reading, just waiting), time to next train, nonverbal response of observer to violation, reaction to dropping the books. After the interaction was completed the team picked up any litter discretely and moved to a different platform. All supervisors and actors were blind to the research hypotheses.

The first survey was administered in the same location as the experiment and parallel to it, in May 2015. The second survey was administered also in the same location in June 2016.

Intensity of direct punishment

The supervisors were instructed to precisely record the form of direct punishment by observers (if any), immediately after punishment occurred. We use these records to classify the intensity (severity) of direct punishment. We recruited two German-speaking research assistants who were unrelated to our study and blind to the hypotheses. They were asked to independently classify each reported instance of punishment as being of either ‘low’, ‘medium’ or ‘high’ intensity. After producing independent ratings, the two assistants met and discussed cases of disagreement, in order to obtain a final coding. The Spearman rank correlation coefficient between their respective ratings was 0.52 ( P <0.001).

Figure 3 shows the incidences of low, medium and high intensity of direct punishment for small and large violations, using the final coding. The difference is not statistically significant (χ 2 (2)=1.024, P =0.60, N =59). We note that our findings regarding the statistically indistinguishable intensity of direct punishment across treatments are robust when we consider each assistant’s pre-agreement coding separately (Assistant 1: χ 2 (2)=1.136, P =0.57, N =59; Assistant 2: χ 2 (2)=0.431, P =0.81, N =59).

Power calculations and sample size

Convention prescribes that the sample size is such that a given treatment effect will be significant at the 5% level, 80% of the time (that is, the power of the test is 80%). Our study is the first to explore the relationship between the severity of norm violation and punishment in a natural field setting. We therefore have no prior on the potential size of the treatment effect (if any). To ensure that our statistical tests are sufficiently powered we used the following approach to calculate our sample size.

As a benchmark for the expected rate of direct punishment in treatment `Small No Help' we use the direct punishment rate that we observed for women in treatment BasePun from Balafoutas et al . 20 where the violation and the experimental location were the same as in our study. That rate was 26.3%. In order to have an estimate of the expected rate of direct punishment with the large violation, we refer to observed differences in punishment rates between small and large violations in the two lab studies that introduced altruistic punishment in the literature, using a similar population. In Fehr and Gächter 14 individuals are willing to invest more than twice the amount they do for ‘small violations’ in order to punish ‘large violations’: taking into account the non-linear punishment technology, comparing small violations (that is, negative deviations of at most 8 tokens of the violator’s contribution to the public good from others’ average contributions) to large violations (that is, negative deviations of more than 8 points) reveals that the amount spent on punishment is on average about 2.25 times greater for large compared with small violations (Fehr and Gächter 14 , p. 991). The same calculations using the data reported in Fehr and Gächter 13 lead to a factor of between 2.30 and 3.00. (As we show in section `Negative emotions in response to norm violations' below, Fehr and Gächter 13 also use these ranges to distinguish between ‘small violations’ and ‘large violations’. We also show that the large violation in our experiment triggers a similar increase in negative emotions.) We take a conservative approach and use the smallest of these factors, 2.25. This leads to an expected punishment rate of (26% × 2.25=58.5%) in treatment `Large No Help'.

To minimize the likelihood of a Type-II error in case the treatment differences turned out to be smaller in the field (that is, failing to reject a false null hypothesis of no differences), we demanded a power of at least 99%. That is, if the rate of direct punishment in the field is as sensitive to the relative severity of the violation as in the lab, then we would be almost certain to detect it at the 5% level. If our statistical results fail to detect a significant difference, therefore, this could not be attributed to our tests being underpowered. Given the aforementioned punishment rates, the sample size needed to identify significant treatment differences at the 5% level with a power of 0.99 is estimated to be at least 160 observations in total, or N =80 per violation. We aimed to collect approximately 100 observations per violation giving us a power of 99.8% given the assumptions above.

As it turned out, in our field experiment, we observe 18.63% of Observers directly punishing a norm violator in `Small No Help'. The difference between the rate of 26.3% (15 out of 57 women) and 18.6% (19 out of 102) is not statistically significant (χ 2 (1)=1.286, P =0.257). If we had used the direct punishment rate of 18.6% for our power analysis, all else equal, we would expect a punishment rate of 41.85% (=18.6% × 2.25) in `Large No Help'. With a sample size of 100 in each treatment, this would imply a power of 96%.

Negative emotions in response to norm violations

Fehr and Gächter 13 argue that the punishment of strangers is likely to be propelled by the negative emotions triggered by the violation of social norms: “Free riding may cause strong negative emotions among the cooperators and these emotions, in turn, may trigger their willingness to punish the free riders. If this conjecture is correct, we should observe particular emotional patterns in response to free riding’’ (p. 139). In this section we explore whether the large violation in our experiment triggered stronger negative emotions than the small violation, and how any difference compares to that reported in Fehr and Gächter 13 . The study of Fehr and Gächter may be the best lab study to compare our findings with as they also investigate one-shot interactions between perfect strangers in a similar population, and the punishment of free riders benefits only other individuals in the future.

In Survey 1, respondents were asked to state how much they would be bothered by either the large or the small violation (question 2). Assigning numbers to each category (0=Not at all, 1=Yes—a little, 2=Yes—quite a lot, 3=Yes—a lot), the weighted average is 2.194 for the large violation, and 1.964 for the small violation. This amounts to an 11.7% increase in negative emotions triggered by the large violation, which is statistically significant (Mann–Whitney U, z =−3.059, P =0.002, N =510).

This difference is very similar in magnitude to that reported in Fehr and Gächter 13 . Subjects in that study had to indicate how annoyed they would be by a group member in a public good game who contributed either 3 (small violation) or 14 (large violation) units less than his/her peers on average, using a seven-point scale (1=’not at all’ to 7=‘very much’). (Notice that these numbers are consistent with our definitions of small and large violations in the previous section.) Fehr and Gächter 13 do not report the average response in their paper but they do write that: “It turns out that a free rider triggered much anger among the other subjects if these subjects contributed a lot relative to the free rider (scenario 1). Forty-seven per cent of the subjects indicated an anger level of 6 or 7 and another 37% indicated an anger level of 5. If the deviation of the free rider’s contribution from the other members’ contribution was relatively small (scenario 2), the anger level was significantly lower (Wilcoxon signed rank test, z=9.636, P<0.0005) but still considerable. In this case (scenario 2), 17.4% of the subjects indicated an anger level of 6 or 7 and 80.5% indicated a level of 4 or 5 in scenario 2’’.

Using the data from Fehr and Gächter 13 , we find that for large violations (scenario 1) and for small violations (scenario 2) the average responses were 5.49 and 4.77 respectively. This amounts to a 15% increase, which is similar to the increase in negative emotions in our experiment (11.7%). Thus, a similar increase in negative emotions to that observed in our experiment led to a threefold increase in punishment in Fehr and Gächter 13 .

Data availability

All relevant data are available from the authors.

Additional information

How to cite this article: Balafoutas, L. et al . Altruistic punishment does not increase with the severity of norm violations in the field. Nat. Commun. 7, 13327 doi: 10.1038/ncomms13327 (2016).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Nowak, M. A. Five rules for the evolution of cooperation. Science 314 , 1560–1563 (2006).

Article   CAS   ADS   Google Scholar  

Nowak, M. A. Evolving cooperation. J. Theor. Biol. 299 , 1–8 (2012).

Article   MathSciNet   Google Scholar  

Guala, F. Reciprocity: weak or strong? What punishment experiments do (and do not) demonstrate. Behav. Brain Sci. 35 , 1–15 (2012).

Article   ADS   MathSciNet   Google Scholar  

Andreoni, J. Reasonable doubt and the optimal magnitude of fines: should the penalty fit the crime? RAND J. Econ. 22 , 385–395 (1991).

Article   Google Scholar  

Boyd, R., Gintis, H., Bowles, S. & Richardson, P. J. The evolution of altruistic punishment. Proc. Natl Acad.Sci. USA 100 , 3531–3535 (2003).

Bowles, S. & Gintis, H. The evolution of strong reciprocity: cooperation in heterogeneous populations. Theor. Popul. Biol. 65 , 17–28 (2004).

Hauert, C., Traulsen, A., Brandt, H., Nowak, M. A. & Sigmund, K. Via freedom to coercion: the emergence of costly punishment. Science 316 , 1905–1907 (2007).

Article   CAS   ADS   MathSciNet   Google Scholar  

Henrich, J. & Boyd, R. Why people punish defectors: weak conformist transmission can stabilize costly enforcement of norms in cooperative dilemmas. J. Theor. Biol. 208 , 79–89 (2001).

Article   CAS   Google Scholar  

Bowles, S., Choi, J.-K. & Hopfensitz, A. The co-evolution of individual behaviors and social institutions. J. Theor. Biol. 223 , 135–147 (2003).

Gintis, H., Smith, E. A. & Bowles, S. Costly signaling and cooperation. J. Theor. Biol. 213 , 103–119 (2001).

Article   CAS   MathSciNet   Google Scholar  

Eldakar, O. T., Farrell, D. L. & Wilson, D. S. Selfish punishment: altruism can be maintained by competition among cheaters. J. Theor. Biol. 249 , 198–205 (2007).

Nakamaru, M. & Dieckmann, U. Runaway selection for cooperation and strict-and-severe punishment. J. Theor. Biol. 257 , 1–8 (2009).

Fehr, E. & Gächter, S. Altruistic punishment in humans. Nature 415 , 137–140 (2002).

Fehr, E. & Gächter, S. Cooperation and punishment in public goods experiments. Am. Econ. Rev. 90 , 980–994 (2000).

Masclet, D., Noussair, C., Tucker, S. & Villeval, M. C. Monetary and nonmonetary punishment in the voluntary contributions mechanism. Am. Econ. Rev. 93 , 366–380 (2003).

Gächter, S., Renner, E. & Sefton, M. The long-run benefits of punishment. Science 322 , 1510 (2008).

Article   ADS   Google Scholar  

Gürerk, Ö., Irlenbusch, B. & Rockenbach, B. The competitive advantage of sanctioning institutions. Science 312 , 108–111 (2006).

Sutter, M., Haigner, S. & Kocher, M. Choosing the stick or the carrot?—Endogenous institutional choice in social dilemma situations. Rev. Econ. Stud. 77 , 1540–1566 (2010).

Balafoutas, L. & Nikiforakis, N. Norm enforcement in the city: a natural field experiment. Eur. Econ. Rev. 56 , 1773–1785 (2012).

Balafoutas, L., Nikiforakis, N. & Rockenbach, B. Direct and indirect punishment among strangers in the field. Proc.Natl Acad.Sci. USA 111 , 15924–15927 (2014).

Denant-Boemont, L., Masclet, D. & Noussair, C. Punishment, counterpunishment and sanction enforcement in a social dilemma experiment. Econ. Theor. 33 , 145–167 (2007).

Nikiforakis, N. Punishment and counter-punishment in public good games: Can we really govern ourselves? J. Public Econ. 92 , 91–112 (2008).

Nikiforakis, N., Noussair, C. & Wilkening, T. Normative conflict and feuds: The limits of self-enforcement. J. Public Econ. 96 , 797–807 (2012).

Fehl, K., Sommerfeld, R. D., Semmann, D., Krambeck, H.-J. & Milinski, M. I dare you to punish me—Vendettas in games of cooperation. PLoS ONE 7 , e45093 (2012).

Wolfgang, M. E., Figlio, R. M., Tracy, P. E. & Singer, S. I. The National Survey of Crime Severity Department of Justice, Bureau of Justice Statistics (1985).

Dreber, A., Rand, D. G., Fudenberg, D. & Nowak, M. A. Winners don’t punish. Nature 452 , 348–351 (2008).

Herrmann, B., Thöni, C. & Gächter, S. Antisocial punishment across societies. Science 319 , 1362–1367 (2008).

Boyd, R., Gintis, H. & Bowles, S. Coordinated contingent punishment is group-beneficial and can proliferate when rare. Science 328 , 617–620 (2010).

Sigmund, K., De Silva, H., Traulsen, A. & Hauert, C. Social learning promotes institutions for governing the commons. Nature 466 , 861–863 (2010).

Traulsen, A., Röhl, T. & Milinski, M. An economic experiment reveals that humans prefer pool punishment to maintain the commons. Proc. R. Soc. B 279 , 3716–3721 (2012).

Rand, D. G., Armao, J., Nakamaru, M. & Ohtsuki, H. Anti-social punishment can prevent the co-evolution of punishment and cooperation. J. Theor. Biol. 265 , 624–632 (2010).

Rand, D. G. & Nowak, M. A. The evolution of antisocial punishment in optional public goods games. Nat. Commun. 2 , 434 (2011).

Hauser, O. P., Nowak, M. A. & Rand, D. G. Punishment does not promote cooperation under exploration dynamics when anti-social punishment is possible. J. Theor. Biol. 360 , 163–171 (2014).

Janssen, M. A. & Bushman, C. Evolution of cooperation and altruistic punishment when retaliation is possible. J. Theor. Biol. 254 , 541–545 (2008).

Download references

Acknowledgements

We thank the research assistants Jarid Zimmermann, Jenna Strzykala, Karen Heuermann, Marcin Waligora, Sebastian Schneiders, and Anne Schielke and the actors Christina Woike, Millie Vikanis, Mirjam Piesker-Muhl, Silvana Lepsa, and Verena Volland for their excellent support. We thank Simon Gächter for sharing the data from Fehr and Gächter (2002). We also thank Ernst Fehr, Manfred Milinski, David Rand, Matthias Sutter, and participants at the University of Cologne seminar, New York University Abu Dhabi seminar, the 1st NYU Global Network Workshop in Experimental Social Sciences, and the 85th Annual Meetings of the Southern Economic Association in New Orleans for useful comments.

Author information

Authors and affiliations.

Department of Public Economics, University of Innsbruck, Universitätsstraße 15, Innsbruck, A-6020, Austria

Loukas Balafoutas

Division of Social Science, New York University Abu Dhabi, P.O. Box 129188, Abu Dhabi, UAE

• Nikos Nikiforakis

Department of Economics, University of Cologne, Universitätsstraße 22a, Köln, D-50923, Germany

Bettina Rockenbach

You can also search for this author in PubMed   Google Scholar

Contributions

L.B., N.N. and B.R. designed the experiment, prepared the surveys, and wrote the manuscript, L.B. compiled the dataset and carried out the statistical analysis, B.R. trained and supervised the team of research assistants.

Corresponding author

Correspondence to Bettina Rockenbach .

Ethics declarations

Competing interests.

The authors declare no competing financial interests.

Supplementary information

Supplementary information.

Supplementary Note 1, Supplementary Methods and Supplementary References (PDF 571 kb)

Peer Review File (PDF 248 kb)

Rights and permissions.

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Reprints and permissions

About this article

Cite this article.

Balafoutas, L., Nikiforakis, N. & Rockenbach, B. Altruistic punishment does not increase with the severity of norm violations in the field. Nat Commun 7 , 13327 (2016). https://doi.org/10.1038/ncomms13327

Download citation

Received : 06 April 2016

Accepted : 22 September 2016

Published : 01 November 2016

DOI : https://doi.org/10.1038/ncomms13327

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

This article is cited by

Group environment promotes the third-party punishment for maintaining social fairness: evidence from erps and neural oscillations.

• Yingjie Liu

Current Psychology (2024)

Experimental evidence shows that ulterior motive attribution drives counter-punishment

• Manuel Muñoz-Herrera

Journal of the Economic Science Association (2023)

A Naturalistic Study of Norm Conformity, Punishment, and the Veneration of the Dead at Texas A&M University, USA

• Michael Alvard
• Katherine Daiy

Human Nature (2021)

Direct and indirect punishment of norm violations in daily life

• Catherine Molho
• Joshua M. Tybur
• Daniel Balliet

Nature Communications (2020)

The benefit of the doubt: willful ignorance and altruistic punishment

• Robert Stüber

Experimental Economics (2020)

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.

Altruism in Its Personal, Social, and Cultural Contexts: An Introduction

• First Online: 01 January 2013

Cite this chapter

• Juris G. Draguns 2  

Part of the book series: International and Cultural Psychology ((ICUP))

2006 Accesses

5 Citations

Altruism is a complex, multifaceted concept that is difficult to define and challenging to investigate. At its core, it refers to intrinsically motivated action for the benefit of other human beings. Proponents of universal egoism attempt to reduce altruism to a surface manifestation of more fundamental, self-gratifying motives. Research over the last several decades has resulted in a partial paradigm shift toward the conclusion that altruism cannot be reduced to egotistic sources and that its roots are deeply embedded in human nature. A variety of research methods have been developed and are being vigorously pursued. They range from studies of altruistic behavior in social psychology experiments to naturalistic field and biographical approaches, sometimes supplemented by psychometric methods. Researchers have studied in exceptional situations as well as in daily lives and have introduced innovative approaches for its exploration. Several sources of altruistic motivation and conduct have emerged, ranging from biological to sociocultural. Yet a great many questions remain to be answered by multimethod and interdisciplinary inquiry before the biological, psychological, and cultural threads are integrated into a scientifically based understanding of human potential for altruism in its manifold modes of expression.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Altruism (See Authenticity)

Six Sources of Altruism: Springs of Morality and Solidarity

Altruism, Morality, and the Morality of Altruism

Bar-Tal, D. (1982). Sequential development of helping behavior: A cognitive-learning approach. Developmental Review, 2 , 101–102.

Article   Google Scholar  

Batson, C. D. (1991). The altruism question: Toward a social-psychological answer . Hillsdale, NJ: Erlbaum.

Google Scholar  

Batson, C. D. (2010). Empathy-induced altruistic motivation. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior (pp. 15–35). Washington, DC: American Psychological Association.

Chapter   Google Scholar  

Batson, C. D. (2011). Altruism in humans . New York: Oxford University Press.

Becker, S. W., & Eagly, A. H. (2004). The heroism of men and women. American Psychologist, 59 , 163–178.

Article   PubMed   Google Scholar  

Bierhoff, H. W. (2001). Prosocial behavior . Hove, East Sussex: Psychology Press.

Bierhoff, H. W., Klein, R., & Kramp, P. (1991). Evidence for the altruistic personality from data on accident research. Journal of Personality, 59 , 263–280.

Boehnke, K. (1988). Prosoziale Motivation, Selbstkonzept und politische Orientierung: Entwicklungsbedingungen im Jugendalter [Prosocial motivation, self concept, and political orientation: Developmental conditions of change in adolescence] . Frankfurt: Peter Lang.

Boehnke, K., Silbereiseen, R. K., Eisenberg, N., Reykowski, J., & Palmonari, A. (1989). Developmental pattern of prosocial motivation: A cross-national study. Journal of Cross-­Cultural Psychology, 20 , 219–243.

Bond, M. H., et al. (2004). Culture-level dimensions of social axioms and their correlates across 41 nations. Journal of Cross-Cultural Psychology, 35 , 548–570.

Buss, D., et al. (1990). International preferences in selecting mates: A study of 37 cultures. Journal of Cross-Cultural Psychology, 21 , 5–47.

Carlo, G., Koller, S. H., Eisenberg, N., Da Silva, M. S., & Frohlich, C. B. (1996). A cross-national study of the relation among prosocial moral reasoning, gender role orientation, and prosocial moral reasoning, gender role orientation, and prosocial behaviors. Developmental Psychology, 32 , 231–240.

Cialdini, R. B., Brown, S. L., Lewis, B. P., Luce, C., & Neuburg, S. L. (1997). Reinterpretation of the empathy-altruism relationship: When one into one equals oneness. Journal of Personality and Social Psychology, 52 , 749–758.

Claar, A., Boehnke, K., & Silbereisen, R. K. (1984). Zur Entwicklung von Motiven prosozialen Handelns bei 12- bis 18-jaehrigen deutschen und polnischen Schuehlern [Development of motives of prosocial action among 12 to 18 year old German and Polish students]. In A. Strikrud (Ed.), Jugend und Werte –Aspekte einer politischen Psychologie des Jugendalters (pp. 166–183). Weinheim: Beltz.

Draguns, J. G. (2007). Empathy across national, cultural, and social barriers. Baltic Journal of Psychology, 8 , 5–20.

Eisenberg, N., Boehnke, K., Schuhler, K., & Silbereisen, R. K. (1985). Development of prosocial behavior and cognition in German children. Journal of Cross-Cultural Psychology, 16 , 69–82.

Eisenberg, N., Carlo, G., Murphy, B., & Van Court, P. (1995). Prosocial development in late adolescence: A longitudinal study. Child Development, 66 , 1179–1197.

Eisenberg, N., Cumberland, A., Guthrie, I. K., Murphy, B. C., & Shepard, S. A. (2005). Age changes in prosocial responding and moral reasoning in adolescence and early adulthood. Journal of Research on Adolescence, 15 , 235–260.

Eisenberg, N., & Miller, P. (1987). Empathy and prosocial behavior. Psychological Bulletin, 101 , 91–119.

Eisenberg, N., & Strayer, J. (Eds.). (1987). Empathy and its development . New York: Wiley.

Feldman, R. E. (1968). Response to compatriot and foreigner who seek assistance. Journal of Personality and Social Psychology, 10 , 202–214.

Gilbert, M. (2003). The righteous: The unsung heroes of the Holocaust . New York: Holt.

Hallie, P. P. (1979). Lest innocent blood be shed: The story of the village of Le Chambon and how goodness happened there . New York: Harper and Row.

Hein, G., & Singer, T. (2010). Neuroscience meets social psychology: An integrative approach to human empathy and prosocial behavior. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior (pp. 109–126). Washington, DC: American Psychological Association.

Hoffman, M. I. (1981). Is altruism part of human nature? Journal of Personality and Social Psychology, 40 , 121–137.

Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across cultures (2nd ed.). Thousand Oaks, CA: Sage.

Hunt, M. (1990). The compassionate beast: What science is discovering about the humane side of humankind . New York: William Morrow and Co.

Kakavoulis, A. (1998). Early childhood altruism: How parents see prosocial behavior in their young children. Early Child Development and Care, 140 , 115–126.

Knafo, A., & Israel, S. (2010). Genetic and environmental influences on prosocial behavior. In M. Mikulincer & P. Shaver (Eds.), Prosocial motives, emotions, and behavior (pp. 149–169). Washington, DC: American Psychological Association.

Kraus, M. W., Piff, P. K., Mendoza-Denton, R., Rheinschmidt, M. L., & Keltner, D. (2012). Social class, solipsism, and contextualism: How the rich are different from the poor. Psychological Review, 119 , 546–572.

Krebs, D. L. (1975). Empathy and altruism. Journal of Personality and Social Psychology, 32 , 1134–1146.

London, P. (1970). The rescuers: Motivational hypotheses about Christians who saved the Jews from the Nazis. In J. Macauley & L. Berkowitz (Eds.), Altruism and helping behavior (pp. 241–250). New York: Academic Press.

Maner, J. K., Luce, C. L., Neuberg, S. L., Cialdini, R. B., Brown, S., & Sagarin, B. J. (2002). The effects of perspective taking on motivation for helping: Still no evidence for altruism. Personality and Social Psychology Bulletin, 28 , 1601–1610.

McCauley, C., & Bock, J. G. (2004). Why does violence trump peace building?: Negativity bias in intergroup relations. In Y. T. Lee, C. McCauley, F. Mogghaddam, & S. Worchel (Eds.), The psychology of ethnic and cultural conflict (pp. 273–288). Westport, CT: Praeger.

McCrae, R. R., & Allik, J. (Eds.). (2002). The five factor model of personality across cultures . New York: Kluwer Academic/Plenum. Mehrabian, A., & Epstein, N. (1972). A measure of emotional empathy. Journal of Personality, 40, 525–543.

Mikulincer, M., & Shaver, P. R. (2005). Attachment security, compassion, and altruism. Current Directions in Psychological Science, 14 , 33–38.

Monroe, K. R. (1996). The heart of altruism: Perception of a common reality . Princeton, NJ: Princeton University Press.

Monroe, K. R. (2002). Explicating altruism. In S. G. Post, G. Underwood, J. P. Schloss, & W. P. Hurlbut (Eds.), Altruism & altruistic love: Science, philosophy, and religion in dialogue (pp. 106–122). New York: Oxford University Press.

Oliner, S. P. (2002). Extraordinary acts of ordinary people: Faces of heroism and altruism. In S. G. Post, G. Underwood, J. P. Shloss, & W. B. Hurlbut (Eds.), Altruism & altruistic love: Science and religion in dialogue (pp. 123–139). New York: Oxford University Press.

Oliner, S. P., & Oliner, P. M. (1988). The altruistic personality: Rescuers of Jews in Nazi Europe . New York: Free Press.

Penner, P. S. (1995). Altruistic behavior: An inquiry into motivation . Amsterdam: Rodopi.

Piliavin, J. A., & Charng, H. V. (1990). Altruism: A review of recent theory and research. American Sociological Review, 16 , 27–65.

Post, S. G. (2003). Profiles of unlimited love: Lives ennobled by purpose. In S. G. Post, B. Johnson, M. E. McCollough, & J. P. Schloss (Eds.), Research on altruism & love (pp. 183–215). Philadelphia: Templeton Foundation Press.

Rohner, R. P. (2004). The parental acceptance-rejection syndrome. American Psychologist, 59 , 830–849.

Royzman, E. & Kumar, B. (2001). On the relative preponderance of empathetic sorrow and its relation to commonsense morality. New Ideas in Psychology, 19, 131–144.

Rushton, J. R. (1984). The altruistic personality: Evidence from laboratory, naturalistic, and self-­report. In E. Staub, D. Bar-Tal, J. Karylowski, & J. Reykowski (Eds.), The development and maintenance of prosocial behavior: International perspectives on positive morality (pp. 271–290). New York: Plenum Press.

Rushton, J. R., Fulker, D. W., Neal, M. C., Nias, D. K., & Eysenck, H. J. (1986). Altruism and aggression: The heritability of individual differences. Journal of Personality and Social Psychology, 50 , 1192–1198.

Schmitt, D. P., et al. (2004). Patterns and universals of romantic attachment across 62 regions. Journal of Cross-Cultural Psychology, 26 , 92–116.

Schwartz, S. H. (2010). Basic values: How they motivate and inhibit prosocial behavior. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior (pp. 221–241). Washington, DC: American Psychological Association.

Silbereisen, R., Lamsfuss, S., Boehnke, K., & Eisenberg, N. (1991). Developmental patterns and correlates of prosocial motives in adolescence. In L. Montada & H. W. Bierhoff (Eds.), Altruism in social systems (pp. 82–104). Lewiston, NY: Hogrefe.

Sommerfeld, E. (2010). The subjective experience of generosity. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior (pp. 303–324). Washington, DC: American Psychological Association.

Sorokin, P. A. (1950). Altruistic love: A study of American “good neighbors” and Christian saints . Boston: Beacon Press.

Staub, E. (2003). The psychology of good and evil: Why children, adults, and groups help and harm others . New York: Cambridge University Press.

Book   Google Scholar  

Stotland, E., Mathews, K. E., Sherman, S. E., Hanson, R. O., & Richardson, B. Z. (1978). Empathy, fantasy, and helping . Beverly Hills, CA: Sage.

Svetlova, M., Nichols, S. R., & Brownell, C. A. (2010). Toddlers’ prosocial behavior: From instrumental to empathetic to altruistic helping. Child Development, 81 , 1814–1827.

Tec, N. (1986). When light pierced the darkness . New York: Oxford University Press.

Tolmacz, R. (2010). Forms of concern: A psychoanalytic perspective. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior (pp. 93–108). Washington, DC: American Psychological Association.

Vaillant, G. E. (1977). Adaptation to life . Boston: Little, Brown.

Vollhart, J. R., & Staub, E. (2011). Inclusive altruism born of suffering: The relationship between adversity and prosocial attitudes and behavior toward disadvantaged outgroups. The American Journal of Orthopsychiatry, 81 , 307–315.

Zahn-Wexler, C., Radke-Yarrow, M., Wagner, E., & Chapman, M. (1992). Developmental concern for others. Developmental Psychology, 28 , 126–136.

Download references

Author information

Authors and affiliations.

Psychology Department, Pennsylvania State University, University Park, PA, 16802-1515, USA

Juris G. Draguns

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Juris G. Draguns .

Editor information

Editors and affiliations.

California Institute of Integral Studies, SETI Institute, Center for SETI Research, Mountain View, 94043, California, USA

Douglas A. Vakoch

Rights and permissions

Reprints and permissions

Copyright information

© 2013 Springer Science+Business Media New York

About this chapter

Draguns, J.G. (2013). Altruism in Its Personal, Social, and Cultural Contexts: An Introduction. In: Vakoch, D. (eds) Altruism in Cross-Cultural Perspective. International and Cultural Psychology. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-6952-0_1

Download citation

DOI : https://doi.org/10.1007/978-1-4614-6952-0_1

Published : 19 April 2013

Publisher Name : Springer, New York, NY

Print ISBN : 978-1-4614-6951-3

Online ISBN : 978-1-4614-6952-0

eBook Packages : Behavioral Science Behavioral Science and Psychology (R0)

Share this chapter

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

• Publish with us

Policies and ethics

• Find a journal
• Track your research

ORIGINAL RESEARCH article

Helping others, warming yourself: altruistic behaviors increase warmth feelings of the ambient environment.

• 1 School of Psychological and Cognitive Sciences and Beijing Key Laboratory of Behavior and Mental Health, Peking University, Beijing, China
• 2 College of Education, Shanghai Normal University, Shanghai, China
• 3 School of Business Administration, Capital University of Economics and Business, Beijing, China

Altruistic behaviors typically improve the welfare of the recipient at the cost of the performer’s resources and energy. Do altruistic performers obtain any positive internal reward from altruistic behaviors? We conducted six experiments to explore whether altruistic behaviors could increase performer’s warmth perception of the ambient environment. The first three studies focused on crisis situations. A retrospective field study (Study 1, with Hurricane Sandy) and two laboratory studies (Studies 2a and 2b, with an earthquake scenario) found that people who helped others felt warmer of the ambient environment than people who did not. We extended to daily life situations and found that participants who performed helping behaviors in laboratory (either voluntarily in Study 3a or randomly assigned to in Study 3b) and passers-by who donated to a charity (Study 4) reported warmer perception of the ambient environment than those who did not. These findings suggested an immediate internal reward of altruism.

Introduction

One who watereth will himself be watered.

Proverbs 11:25

When we talk about altruistic behaviors, we often talk about sacrifice and the potential costs and risks associated with it. At the individual level, altruism is often non-economic and even maladaptive to survival because the altruistic performer needs to share his/her own resources and energy with others without receiving explicit returns ( Batson, 1991 ; Myers, 1993 ; de Waal, 2008 ). When disasters signal a shortage of resources, it seems unwise to behave altruistically. However, at many times, we quote the saying the roses in her hand, the flavor in mine to encourage more altruistic behaviors. Is this quote simply a fortune cookie comment? Could those who behave altruistically literally experience the “flavor” or other positive physical feelings, as implied in this quote? In the current research, we explored the effects of altruistic behaviors on individuals’ physical feelings, focusing in particular on warmth. Warmth is a fundamental need to humans and other primates (e.g., Harlow, 1958 ; IJzerman et al., 2015a ). Moreover, warmth perception of the ambient environment is a typical variable that links the psychological and physical worlds. Specifically, we found that altruistic behaviors could lead the altruistic performer to increase his/her warmth perception of the ambient environment.

The Functional Adaptabilities of Altruistic Behaviors

Researchers have tried to recognize the functional adaptabilities of altruistic behaviors despite the self-sacrificing nature of such behaviors. The kin selection theory ( Hamilton, 1963 ; Hamilton and Axelrod, 1981 ) suggests that altruistic behaviors toward those with shared genes (i.e., offspring or relatives of the altruistic performer) could maximize genetic frequencies at the group level. The reciprocal motivation and social exchange theory suggests that an altruistic performer could expect future returns either directly from the recipient ( Trivers, 1971 ) or indirectly from a third party ( Nowak and Sigmund, 1998 , 2000 ). The above two theories both focus on the long-term benefits for the altruistic performers ( Gintis et al., 2003 ).

However, researchers have uncovered the phenomenon of altruistic punishment. Individuals chose to incur great cost to punish norm violators in a group (e.g., Henrich et al., 2006 ). Such choices were regarded as altruistic for they aimed at restoring fairness and protecting the group norm. Because there was usually no kin relative in the group and the research employed a one-shot game ( Gintis, 2000 ; Boyd et al., 2003 ; Gintis et al., 2003 ; Henrich et al., 2006 ), this phenomenon could not be explained by the two previously mentioned theories. Moreover, a neuroscientific study on altruistic punishment found that effective punishment activated the dorsal striatum. Participants with stronger activation in this brain area were willing to perform more punishment ( de Quervain et al., 2004 ). As the dorsal striatum is an important part of the reward system ( Knutson et al., 2000 ; Delgado et al., 2003 ), the findings implied that altruistic punishment could trigger immediate positive experiences for altruistic performers. This finding was consistent with earlier notions that altruistic behaviors would promote the release of endogenous opioid peptides ( Danielli, 1980 ), which contributed to the control of pain ( Basbaum and Fields, 1984 ) and the modulation of human mood and feelings of well-being ( Leknes and Tracey, 2008 ). Furthermore, these findings implied that we may focus on the internal rewarding system of human altruistic behaviors. Rather than uncover tangible returns as a result of altruistic behaviors, we would like to explore the potential positive effects of altruism on one’s psychological and physical experiences.

Researchers have found positive psychological consequences of altruistic behaviors. For example, prosocial spending including charity donations and gift giving were found to evoke happiness ( Dunn et al., 2008 ). Researchers in this field used the term “warm glow” to indicate an internal sense of satisfaction in donors after donating money ( Harbaugh, 1998 ). Altruistic behaviors also promoted self-efficacy in the elderly ( Midlarsky and Kahana, 1994 ) and enhanced positive self-evaluations ( Post, 2005 ). In the current study, we focused on the warmth perception of the ambient environment. We chose this variable for three reasons. First, warmth is a fundamental need of humans and other primates. This has been supported by earlier studies on development and attachment (e.g., Harlow, 1958 ) as well as by a recent model of thermoregulation ( IJzerman et al., 2015a ). Feeling of warmth could be a source of security ( Harlow, 1958 ) and individuals could use feeling of warmth as an indicator of social resources ( IJzerman et al., 2015a ). Research on winter depression implied that lack of warmth could be a threat to mental health (e.g., Molin et al., 1996 ). If altruism were to lead to threats against other survival-related resources such as food, compensatory feelings of warmth could be a comfort for individuals who are facing crisis situations, which could be an advantage for survival. Second, it is implied from previous research that the reward of altruism might be complicated. Reward, for instance, might be related to an internal reward system rather than simply tangible resources. Warmth perception of the ambient environment is a variable that links the psychological and physical worlds. It describes individuals’ internal reflections of the surrounding world, which could be an important facet of positive feedback of altruism. Third, coldness is a typical threat in many crises, especially in natural environments. For example, extreme cold weather in winter could cause a large amount of deaths (e.g., Webb, 2014 ; Ward, 2015 ). Coldness is also one of the most important environmental factors in mountain sports accidents ( Chamarro and Fernández-Castro, 2009 ). Compared with non-crisis situations, people are more likely to lack resources (e.g., warm food, warm clothes) in coping with coldness. Moreover, individuals may tend to amplify the threats of the crisis-situations. Such tendency was usually related to other psychological reactions toward disasters such as post-traumatic stress disorder (e.g., Heir et al., 2009 ). And the tendency could also be amplified in a social level ( Kasperson et al., 1988 ). Under extreme circumstances, individuals can do little to change the objective situation. Regulating warmth feelings of the ambient environment is potentially helpful for individuals to feel positively about the situations. We hypothesized that altruistic behaviors would increase the feelings of warmth among altruistic performers.

Altruistic Behaviors Promote Physical Warmth

Researchers have linked feelings of warmth to social behaviors and social cognitions. Recently, IJzerman et al. (2015a) proposed a social thermoregulation model. According to this model, thermoregulation is costly for a single individual; therefore, social interactions (e.g., bodily contact) are vital and economic for animals to maintain proper body temperatures. The model argues that the process of social thermoregulation has shaped high-order social cognition. Earlier theories on grounded cognition ( Barsalou, 1999 , 2008 ; Schubert, 2005 ) have suggested that abstract social cognitions are grounded in interactions with the physical world. Metaphoric models emphasize metaphoric mappings between social cognition and physical perception ( Landau et al., 2010 ). These theories help explain the connections between the social and physical worlds in human beings.

An important line of research has focused on a bidirectional relationship between physical warmth and social warmth. On the one hand, a well-known study conducted by Williams and Bargh (2008) showed that a simple manipulation of physical warmth could lead to favorable interpersonal evaluations and behaviors. In addition, Kang et al. (2011) found that physical warmth increased trusting behavior. Such behaviors were regarded conceptually as interpersonal warmth in social life. On the other hand, other studies have yielded evidence that social-related concepts raised warmth perception. For example, research found that thinking about traits and objects that were positively related to communal concepts raised perceived warmth of the ambient environment ( Szymkow et al., 2013 ; IJzerman et al., 2015b ). Moreover, neuroscientific studies have also supported connection between social warmth and physical warmth. Kang et al. (2011) found insula activation when physical warmth increased trusting behavior. Inagaki and Eisenberger (2013) ’s study also found that social-warmth and physical-warmth conditions overlapped on their activations in ventral striatum and middle insula. This overlap was specific in warmth-related positive feelings.

Furthermore, this social-physical link of warmth has been indirectly supported by the “cold” side of social life. Specifically, Zhong and Leonardelli (2008) found that recalling a past experience of social exclusion resulted in a lower estimate of ambient temperature. A social exclusion manipulation could lower skin temperature, and physical warmth experiences were more desirable and effective in comforting the participants experiencing feelings of loneliness ( Bargh and Shalev, 2012 ; Ijzerman et al., 2012 ).

Through the consistently observed relationship between social behaviors (especially positive or prosocial behaviors that foster social relationships) and feelings of warmth, we expected a similar relationship between altruistic behaviors (as typical prosocial behaviors) and warmth perception. Apart from happiness, self-efficacy or some other psychological states, altruistic behaviors could also lead to some positive consequences that are more physical related. Specifically, we predicted that altruistic behaviors would lead to an increased warmth perception of the ambient environment.

Perceived Social Distance as a Mediator

Apart from the research that confirmed a relationship between social warmth and physical warmth, many studies have identified the effects of social distance and physical distance on warmth feelings. For example, IJzerman and Semin (2010) found that a closer perception of social distance (either induced by physical distance or by similarity manipulation) resulted in a higher estimate of ambient temperature. The effects could be reasoned that there is a similar overlapping of social distance and physical interpersonal distance (e.g., IJzerman and Semin, 2009 ), and that physical warmth is usually related to physical interpersonal distance.

In a recent model, social baseline theory was built on previous theories of attachment and other related findings ( Beckes and Coan, 2011 ). The theory proposes that social relationship and social proximity is actually a baseline for human being because it is energy-saving and risk-reducing. Individuals are actually more activated when social relationships are threatened, and they have a tendency to return to “baseline.” Thus, individuals may incorporate relationship-oriented social behaviors to achieve social proximity. Because of the close link between social distance and warmth perception, social distance could be a proximal bridge between prosocial or relationship-oriented behaviors and warmth perception. Thus, we hypothesized that altruistic behaviors would increase the performers’ warmth perceptions of the ambient environment. Moreover, we predicted that this effect would be mediated by decreasing the perceived social distance between altruists and recipients.

Overview of the Current Research

In the current research, we conducted six studies to test the hypothesis that altruistic behaviors would result in a warmer perception of ambient environment and to additionally test the effect of perceived social distance as a mediator. Study 1 was a retrospective field study conducted in the context of Hurricane Sandy, revealing the relationship between altruistic behaviors and perceived warmth in a crisis. Then, in Studies 2a and 2b, we created a crisis situation in the laboratory to replicate this effect and to test the mediation effect of perceived social distance. In Studies 3a and 3b, we extended such effect and the mediation effect of perceived social distance to daily life situations using laboratory experiments. In Study 4, we conducted a field study to replicate such an effect in a real donation activity. To note, all the studies in the current research were set in a comparatively cold environment. We did not explore how this effect would manifest in a much hotter environment.

Study 1 explored the effect of altruistic behaviors on feelings of warmth in a real crisis context. Hurricane Sandy was the most destructive hurricane of the 2012 Atlantic hurricane season, dramatically shocking 24 states in the United States with its ferocity dating back to 29th October, 2012. After the storm crashed ashore, the rain turned into blizzard conditions along the east coast of the United States. According to news reports, the highest snowfall accumulation was 36 inches (91 cm) ( Kellogg, 2013 ). What was worse, due to the heavy snow, the heating and power systems were cut off for more than 10 days in some places in New York State. The extremely cold weather turned the recovery and reconstruction into chronic suffering. As a result, local residents were suffering the continuous stress of environmental threat in subsequent months. In this study, the sample included participants who were local residents at the time of the crisis and who experienced this crisis first-hand. We hypothesized that participants who recalled an altruistic experience (vs. those who recalled a non-altruistic experience) in the crisis would then have a memory of a warmer ambient environment.

Participants

G ∗ Power 3.1 was used to compute a priori power analyses in the study. According to Cohen’s (1992) suggestion, a power of 0.8 and an effect size that was slightly above moderate level (effect size d = 0.6) were used as the input data. The expected effect size was also consistent with previous research on similar topics (e.g., IJzerman and Semin, 2009 , 2010 ). This resulted in an expected sample size of 72. Seventy-nine local residents from the disaster areas (New Jersey State and New York State) (33 males; M a ge = 29.65 years, SD = 9.41) were recruited on Amazon’s Mechanical Turk website (MTurk 1 ). They were asked to complete an online survey named “Life after Sandy Hurricane” and were each rewarded a $5 Amazon gift-card. The purpose of the survey was described as for exploring how Hurricane Sandy and the following extreme cold weather influence people’s daily life. The survey was launched within 3 months after Hurricane Sandy hit the United States while residents in the disaster areas were still suffering through the chronic recovery and reconstruction. Participants were randomly assigned to the altruistic or non-altruistic group. There were 42 participants in the altruistic group and 37 participants in the non-altruistic group.

Materials and Procedure

Participants were first informed that the survey aimed to explore how Hurricane Sandy and the subsequent extreme weather conditions influenced the local residents. Next, participants were asked to recall an experience during or after Hurricane Sandy and to write it down in detail (using at least 50 words). In the altruistic group, participants were asked to “recall an experience in which he/she did something mainly taking other people’s benefits into consideration (e.g., giving food to a homeless person on the street, offering someone a free-ride, volunteer work, etc.).” In the non-altruistic group, they were asked to “recall an experience in which he/she did something mainly taking his/her own benefits into consideration (e.g., rushing to the supermarket for necessities without thinking about how others may need the goods, refusing to offer a free-ride, ignoring neighbor’s asking for help, etc.).” Afterward, participants were asked to recall their instant feelings of the ambient environment after the experience along a 7-point scale (1 = extremely cold , 7 = extremely warm ). As control variables, the participants were also asked to report their feelings of warmth in the current environment along a 7-point scale (1 = extremely cold , 7 = extremely warm ) in addition to the estimated temperature (in degrees Fahrenheit) of the current environment.

Results and Discussion

The participants recalled warmer ambient environment after the recalled altruistic experience ( M = 4.55, SD = 1.19) than after the recalled non-altruistic experience ( M = 3.35, SD = 1.06), t (77) = 4.68, p < 0.001, d = 1.06. The effect remained significant when we included as covariates the warmth feelings and the estimated temperature of the current environment during the survey, F (1,74) = 19.93, p < 0.001, η 2 = 0.21. Moreover, the warmth feelings or the estimated temperature of the current environment during the survey did not show significant group differences [warmth feelings: M altruistic = 4.40, SD = 1.08, M non-altruistic = 3.97, SD = 1.44, t (77) = 1.52, p = 0.134; environmental temperature: M altruistic = 69.55, SD = 5.99, M non-altruistic = 67.39, SD = 10.07, t (76) = 1.17, p = 0.246].

In summary, Study 1 primarily demonstrated the effect of altruistic behavior on warmth feelings in a retrospective real-world crisis. It is worth noting that the results for warmth feelings of the current environment did not reach significance, although in the same direction as for the remembered environment. This seemed not consistent with previous research when recalling a past experience of social exclusion affected current feelings of warmth (e.g., Zhong and Leonardelli, 2008 ). This was possibly due to individuals’ separation of the current feelings and remembered feelings. In addition, the study was less controlled due to the nature of the real crisis scenario. Thus, well-controlled experiments were conducted in the laboratory by measuring and manipulating altruistic behaviors, and warmth feelings were collected instantly after the behaviors.

Study 2a was designed to replicate the effect of altruistic behavior on warmth perception of the ambient environment in crisis context with three improvements. First, we created an earthquake scene in the laboratory. Participants were asked to imagine that they were stuck in an earthquake scene in which a stranger was requesting help. Second, the temperature of this laboratory-based earthquake scene was kept stable at 15°C. Third, altruistic willingness was directly measured in this study so that we could observe the relationship between altruistic willingness and warmth feelings. To match the settings of the scenario, participants took the experiment in groups of 4 and there was no former acquaintance within each group. Such design could increase a sense of reality and offer a closer simulation of the scenario. It could also be more natural to induce helping behaviors in such conditions. We expected that the participants with higher altruistic willingness would perceive higher levels of warmth of the ambient environment.

Because of the correlational nature of Study 2, a power of 0.8 and a medium level of r ( r = 0.3) were used to compute the expected sample size ( Cohen, 1992 ). This resulted in an expected sample size of 67. Study 2a took place in a university in Beijing, China. We recruited 69 college students (24 males; M a ge = 22.31 years, SD = 2.97) from the campus online forum. All participants read the informed consent document and agreed to participate in the experiment for a payment of 10 RMB (approximately $1.5).

As mentioned before, the procedure was implemented in groups of four participants with no former acquaintances. They were first guided into a preparation room where they read a brief introduction of the experiment and then led to the experiment room. The experiment room was set as a scene after an earthquake, with a constant temperature of 15°C and all lights turned off. Participants were asked to sit in a circle on the floor during the whole experiment. They each were approximately 40–50 cm apart from one another.

After ensuring that each of the four participants was ready, the laptops started synchronously to play a video containing standardized instructions leading participants through the whole experiment. Participants were asked not to interact with other participants. The first part of the video was a 40-s video clip of a real earthquake. Participants were instructed to imagine that they had just experienced the earthquake and were currently stuck in a room under a collapsed building with several tourists.

In the second part, the video displayed a questionnaire with several slides, and participants were asked to write down their responses on the paper. First, they were required to report how was their perception of the ambient environment along an 11-point scale (0 = extremely cold , 10 = extremely warm ). This was served as a pre-altruism measure of warmth perception. Perceived severity and uncertainty of the situation were also measured as control variables. Second, participants were told that “one of the tourists did not have food and had requested help.” Participants were asked to “indicate (with a whole number between 0 and 100) the percentage of food they would like to share with the tourist.” The willingness-to-help was regarded as participants’ subjective willingness of altruistic behavior. Additionally, warmth perception of the ambient environment was again measured as a post-altruism measure. Finally, participants were debriefed and then paid for their participation.

To control for individual differences in feelings of warmth before altruistic behaviors, we subtracted the pre-altruism measure from the post-altruism measure of warmth perception and used this result in analyses. Regression analysis showed that willingness-to-help was positively related to change in warmth perception of ambient environment, F (1,67) = 4.39, p = 0.040, β = 0.25. This relationship remained significant when perceived severity and uncertainty of the scenario were controlled in the first step of the regression, F (1,65) = 4.46, p = 0.039, β = 0.26.

Study 2a replicated an association between altruism and warmth perception of ambient environment, demonstrating that people with a greater subjective altruistic willingness would experience warmer feelings. However, limitations remained in the measurement of altruism. First, neither Study 1 (recalled altruistic behavior) nor Study 2a (altruistic willingness) measured actual altruistic behavior. Second, the results of Study 2a were correlational, so a causal relationship between altruism and warmth feeling was not confirmed. Third, it remained unclear why the effect of altruism on perceived physical warmth emerged.

Study 2b aimed at confirming the causal relationship between altruism and warmth perception in a crisis situation. Participants were randomly manipulated to exhibit an altruistic or a non-altruistic behavior in an experiment. We also explored a possible mediator of the relationship, hypothesizing that participants in the altruistic group (vs. the non-altruistic group) would report a warmer perception of the ambient environment because they felt a closer social distance to the help seeker. Moreover, we measured participants’ prosocial traits to control for the possible confounding effects of individual differences.

The computation of the sample size was identical to that of Study 1, resulting in a sample size of 72. Study 2b took place in a university in Beijing, China. Eighty-five university students were recruited from the campus online forum and randomly assigned to one of the two conditions (altruistic vs. non-altruistic). Five participants were excluded from the final data analysis because they responded to less than two-thirds of the assessment items. The final sample (28 males; M a ge = 22.41 years, SD = 3.04) comprised 36 participants in the altruistic condition and 44 participants in the no-altruistic condition. Participants each received a payment of 10 yuan RMB (approximately $1.5).

The settings and procedure of Study 2b were almost identical to those of Study 2a except for several important changes. First, although participants still participated in groups of four, one of the four participants in each group was actually a confederate.

Second, in addition to reading a brief introduction of the experiment in the preparation room, each participant was given a bag. They were told not to open and examine the bag until the instructions told them to do so in the experiment. The three bags for the real participants each contained ten small packs of bread (20 g each). There was no food in the confederate’s bag.

Third, the participants were guided to the experiment room, which had the same settings as in Study 2a. But after watching the first part of the video, the participants were instructed to open the bag and count the bread packs. Because participants were sitting close to each other (∼40–50 cm away from each other), they could easily notice that the confederate had nothing in the bag.

Fourth, participants were then instructed to report (1) the pre-altruism measure of warmth perception of the ambient environment and (2) control variables. Apart from the perceived severity and uncertainty of the situation, negative emotions (i.e., sadness, fear, desperation, and stress; α = 0.85) were also included as control variables.

Fifth, after the above measures were taken, participants were told that the person with no food in bag had requested help. Moreover, participants were manipulated into two groups by reading the following words: ‘ gathering everyone’s power and sharing resources with the group were the best choice in crisis’ (altruistic group) or ‘ saving one’s own resources and maximizing one’s own benefits were the best choice ’ (non-altruistic group). Each participant was asked to decide how many bread packs he/she would like to give to the help-seeker and to take the corresponding number of packs out of the bag. Participants were also instructed to write down the number on the answer sheet. This was used as the manipulation check of altruistic behavior.

Sixth, participants were asked to report perceived psychological distance with the other tourists in the situation on a 7-point scale (1 = close , 7 = distant ) and this measure of social distance was used as the proposed mediator. The post-altruism measure of warmth perception was measured identically to Study 2a. Finally, participants completed the Social Value Orientation Questionnaire as a measure of prosocial traits after returning to the preparation room.

Prosocial Trait

We used the Social Value Orientation Questionnaire ( Van Lange et al., 1997 ) to measure prosocial traits. This questionnaire contains nine multiple-choice situations. In each situation, a participant needed to choose one of three options to decide the outcomes for himself/herself and for another player. The three options in each situation corresponded to three social value orientations: competitive (seeking a larger difference between one’s own and other’s outcomes), individualistic (seeking a larger outcome for oneself), and prosocial (seeking a larger joint outcome). If the choices of six or more out of the nine situations were consistent with one of these social value orientations, participants were classified accordingly. Otherwise, the participant was designated as unclassified.

The manipulation check of altruistic behavior was successful. Participants in the altruistic condition shared more packs of bread ( M = 3.11, SD = 0.98) with the help-seeker than those in the non-altruistic condition ( M = 2.55, SD = 1.13), t (78) = 2.36, p = 0.021, d = 0.53.

In terms of change in warmth perception of the ambient environment, the result was consistent with Studies 1 and 2a. Participants in the altruistic group reported more increase in warmth feelings of the ambient environment ( M = 0.44, SD = 2.34) than those in the non-altruistic group ( M = -0.70, SD = 2.53), t (78) = 2.09, p = 0.040, d = 0.47. The effect remained significant when negative emotions, perceived severity and uncertainty of the scenario were controlled as covariates in the analysis, F (1,74) = 4.33, p = 0.041, η 2 = 0.06. 2

A Chi-square test showed that there was a tendency that the distributions of the four social value orientations (competitive, individualistic, prosocial, and unclassified) in the two groups were not balance [χ 2 (3) = 6.43, p = 0.092]. We also tested whether social value orientations affect change in warmth perception. Because social value orientation was a categorical variable, we run a 2 (experimental manipulation: altruistic vs. non-altruistic) × 4 (social value orientation: competitive vs. individualistic vs. prosocial vs. unclassified) ANOVA. Because there was no significant interactions between the experimental manipulation and the social value orientation ( p = 0.524), a custom model was run for analyzing only the main effects. Results showed that the effect of experimental manipulation remained marginally significant, F (1,75) = 3.39, p = 0.070, η 2 = 0.04. And social value orientation did not significantly affect change in warmth perception, F (3,75) = 1.50, p = 0.232, η 2 = 0.06.

Next, we examined whether perceived social distance mediated these effects. Participants in the altruistic group reported a significantly shorter perceived distance ( M = 2.53, SD = 1.00) than those in the non-altruistic group ( M = 3.16, SD = 1.46), t (78) = 2.20, p = 0.031, d = 0.50. Regression analysis showed that the group variable (the altruistic group coded as “1” and the non-altruistic group coded as “0”) was a significant predictor of warmth perception (β = 0.23, p = 0.040). However, the strength of this relationship became non-significant (β = 0.18, p = 0.105) when perceived social distance was included in the analysis. Moreover, perceived social distance had a tendency to be negatively related to warmth perception (β = -0.19, p = 0.091). The above results imply a mediation effect of perceived social distance. To further confirm the mediation effect, a 5000-sample bootstrapping analysis was conducted. Results showed that the 95% bias-corrected confidence interval for the indirect effect was [0.03, 0.61], suggesting a significant indirect effect ( MacKinnon et al., 2007 ). Therefore, the results of Study 2b confirmed that engaging in altruistic behaviors can increase the perceived warmth of the ambient environment while also suggesting that reduced social distance could be an internal process.

So, the first three studies focused on the relationship between altruism (recalled altruistic behavior, altruistic willingness, and manipulated altruistic behavior) and perceived warmth of the ambient environment. The scenarios used in these studies were all crisis situations with low environmental temperatures. The temperature setting was important because a cold environment could be an extreme threat to survival during a crisis. Thus, regulating warmth perception of the ambient environment would be of great significance.

In the following three studies, we aimed to extend these findings to non-crisis situations with cold environments. This was mainly for two reasons. First, although we explored altruistic behaviors with either real or laboratory-based crisis situations, the manipulation and observation of altruistic behaviors suffered from some limitations. We could obtain only retrospective data from real crisis, and the manipulation of altruistic behaviors was direct. Second, participants in Studies 2a and 2b took the experiments in groups, which could result in interdependence of data. Although we did not find significant influence of group in Study 2b, following studies are aimed to avoid this problem by changing the experimental design. Third, the situations used in the previous studies were crisis-related, and we would like to confirm the relationship in less-threatening daily situations.

Study 3a was designed to examine the association between altruistic behaviors and perceived warmth of the ambient environment in our daily life. We made two improvements in this study. First, we adopted temperature estimation as an additional index of warmth perception. Second, body temperature was measured as a control variable. We predicted that participants who chose to behave altruistically would report feeling warmer in the environment than those who chose not to help. In this study, we also predicted a mediation effect of perceived social distance.

The computation of the sample size was identical to that of Study 1, resulting in a sample size of 72. Study 3a took place in a university in Beijing, China and participants were college students recruited from the campus online forum. The final sample contained 64 participants (33 males; M a ge = 22.66 years, SD = 2.47). Thirty-two participants who engaged in the altruistic task were labeled the altruistic group and the rest were labeled the non-altruistic group.

The temperature of the experiment room was maintained constantly at 15°C. Seventy-one college students were recruited and completed a 10-min irrelevant decision-making questionnaire for a reward of 10 yuan RMB (approximately $1.5). After participants were paid, they were invited to participate in an additional activity organized by the Student Union of the Department of Psychology, but for no extra reward. Seven participants refused to participate and were excluded from the final analysis. Thus, the final sample contained 64 participants.

The additional activity involved two tasks for children from low-income migrant workers’ families. These children are usually regarded as a disadvantaged group in China. The first task was time-consuming and required considerable attention. Participants needed approximately 10 min to read and revise some educational materials for these children. The second task required participants to just complete a 1-min questionnaire about their understandings on these children. In this questionnaire, a 1-item measure of perceived social distance toward these children was presented along a 7-point scale (1 = extremely close , 7 = extremely distant ) with some filling items. Participants could decide whether to participate in both tasks or in the second task only. Based on their decisions, participants were labeled as altruistic or non-altruistic.

Lastly, participants were asked to finish another small survey entitled “A survey on the Laboratory Environment.” In the survey, participants were required to report their perception of the warmth of the experiment room on an 11-point scale (0 = extremely cold , 10 = extremely warm) . They were also asked to estimate the room temperature in degrees Celsius. As a control, we measured the body temperature of the participants using a non-contact infrared thermometer. Finally, participants were debriefed and thanked for their participation.

As expected, participants in the altruistic group felt the room was warmer ( M = 6.06, SD = 2.18) than those in the non-altruistic group ( M = 4.53, SD = 2.09), t (62) = 2.86, p = 0.006, d = 0.72. This effect remained significant when body temperature was included as a covariate, F (1,61) = 9.31, p = 0.003, η 2 = 0.13. Participants in the altruistic group also reported higher estimates of the room temperature ( M = 16.91, SD = 5.21) than those in the non-altruistic group ( M = 12.58, SD = 4.48), t (62) = 3.57, p = 0.001, d = 0.89. This effect also remained significant when body temperature was included as a covariate, F (1,61) = 12.34, p = 0.001, η 2 = 0.17. Moreover, there was a significantly positive correlation between feeling of warmth and estimate of the room temperature ( r = 0.36, p = 0.002). However, body temperature did not correlate with feeling of warmth ( r = -0.20, p = 0.120) or estimate of the room temperature ( r = 0.05, p = 0.671). We then examined whether perceived social distance mediated the effects of group conditions on warmth perception. Participants in the altruistic group reported a significantly shorter distance ( M = 3.66, SD = 1.54) than those in the non-altruistic group ( M = 4.47, SD = 1.32), t (62) = 2.27, p = 0.027, d = 0.57. Regression analysis showed that the group variable (the altruistic group coded as “1” and the non-altruistic group coded as “0”) was a significant predictor of warmth perception, β = 0.34, p = 0.006). This relationship weakened (β = 0.28, p = 0.024) when perceived social distance was included in the analysis. Moreover, perceived social distance had a tendency to be negatively related to warmth perception (β = -0.21, p = 0.087). The above results implied a mediation effect of perceived social distance. To further confirm the mediation effect, a 5000-sample bootstrapping analysis was conducted. Results showed that the 95% bias-corrected confidence interval for the indirect effect was [0.04, 0.89], suggesting a significant indirect effect ( MacKinnon et al., 2007 ).

In conclusion, the effect of altruistic behavior on perceived warmth and the mediating effect of perceived social distance were verified in a daily life situation. In Study 3b, we imported a stricter control of altruistic behaviors to confirm the effect of altruistic behavior on warmth perception of the ambient environment.

In Study 3b, we randomly assigned participants to an altruistic or a non-altruistic condition. Moreover, a no-task group was added as a control group to rule out the alternative explanation that non-altruistic behavior could lead to a decrease in warmth perception of the ambient environment. We hypothesized that participants in the altruistic group would report higher warmth perception than those in the non-altruistic group and the no-task group. We expected participants in the non-altruistic and the no-task groups to report equal feelings of warmth.

Because Study 3b was designed to be consisted of three groups, a power of 0.8 and an effect size f of 0.35 were used to compute the expected sample size, resulting in a sample size of 81. Study 3b took place in a university in Beijing, China. Eighty-three college students were recruited from the campus online forum. Eight of them did not finish the experiment and were thus excluded from the analysis. The final sample contained 75 participants (36 males; M a ge = 22.2 years, SD = 7.42). Participants were randomly assigned to one of the three conditions, resulting in 25 participants in each group.

Participants were led to the experiment room, the temperature of which was maintained constantly at 15°C. In the altruistic condition, participants were invited to take part in a charity activity called ‘The Love Crossing 4000 Kilometers’ to help the students from a remote town called Jimunai. This town was one of the poorest areas in China, located 4000 km from Beijing. The charity aimed to collect stationery for these students and offer them more opportunities to learn about the outside world. Each participant was invited to write a postcard for a particular student in Jimunai Secondary School to introduce an attraction in Beijing. Participants then were given a 10-RMB (approximately $1.5) cash note and were required to put the note in a charity box. Notably, all of the postcards were sent to the corresponding students, and the money was donated to the school in the name of participants after the experiment.

In the non-altruistic condition, participants were asked to read information about Beijing’s candidature for the 2022 Winter Olympic Games. Participants were told that the laboratory was helping to advocate the attractions in Beijing. They could choose one postcard and write down an introduction about an attraction in Beijing. In the no-task group, participants were assigned no additional task.

All the participants then were asked to complete the survey about the laboratory environment, which was identical to that used in Study 3a. In the survey, participants were required to report their warmth perception of the experiment room along an 11-point scale (0 = extremely cold , 10 = extremely warm ). As a control, we measured the body temperature of the participants using a non-contact infrared thermometer. Finally, all the participants were debriefed, thanked, and each paid 10 RMB (approximately $1.5) for their participation.

To ensure the successful manipulation of altruism, we asked another 78 participants (36 males) to read the scenarios in Study 3b and to then indicate the extent of altruism (‘To what extent do you feel you could help others if you participate in this activity?’ 1 = not at all , 5 = great deal ). Results showed that the participants in the altruistic condition reported more helping ( M = 3.63, SD = 0.75) than those in the non-altruistic condition ( M = 3.18, SD = 0.90), t (76) = 2.42, p = 0.018, d = 0.55. To exclude the possibility that the non-altruistic material would induce thoughts of competition, the participants also indicated perceived competition (‘To what extent do you feel you are competing with others?’ 1 = not at all , 5 = great deal ). No significant difference was found between the participants in the two conditions, p = 0.976. In addition, participants were required to write down 5–10 words that the reading material reminded them. We asked two experimenters who did not know the intention of the study to classify the words. Results showed that participants who read the altruistic scenario wrote down more altruistic-related words ( M = 1.06, SD = 0.17) than those who read the non-altruistic scenario ( M = 0.27, SD = 0.04), t (76) = 9.63, p < 0.001, d = 6.56. Similar results were found for caring-related words ( M altruistic = 1.13, SD = 0.70, M non-altruistic = 0.08, SD = 0.27, t (76) = 8.85, p < 0.001, d = 2.02.

One-way ANOVA revealed that the warmth perception of the participants were significantly different across three conditions, F (2,72) = 4.77, p = 0.011, η 2 = 0.12. As expected, Bonferroni post hoc analysis showed that participants in the altruistic group ( M = 6.84, SD = 1.89) felt significantly warmer in the room than those in the non-altruistic group ( M = 5.32, SD = 2.01) (95% CI of mean difference [0.19, 2.85], p = 0.019). Likewise, participants in the altruistic group felt significantly warmer than those in the no-task group ( M = 5.48, SD = 1.83) (95% CI of mean difference [0.03, 2.69], p = 0.042). However, there was no significant difference in warmth perception between the non-altruistic and no-task groups (95% CI of mean difference [-1.17, 1.49], p = 1.00). Moreover, the effect of condition remained significant when the body temperature of participants was included as a covariate, F (2,71) = 4.55, p = 0.014, η 2 = 0.11. And there was no significant correlation between feelings of warmth and body temperature of the participants, r = 0.08, p = 0.490.

With stronger causal inference in this instance, the results replicated the effect of altruism on the warmth perception of the ambient environment. By adding a no-task control group, this study showed that the effects on warmth perception was caused by an increase in warm feelings from behaving altruistically rather than a decrease from not behaving altruistically.

Studies 3a and 3b replicated in the laboratory the primary findings of Studies 2a and 2b with regard to daily life situations. Compared with crisis situations, daily situations were more naturally created. In Study 3a, altruistic behaviors were chosen voluntarily by participants whereas in Study 3b, participants were randomly led to believe that they did an altruistic behavior or a non-altruistic behavior. The results provided convergent evidence for the proposed effect that altruistic behaviors would increase the perception of ambient warmth. To note, Studies 3a and 3b were underpowered by 8 and 6 participants compared to the computed sample sizes, respectively. This was due to experimental constraints of unexpected loss of participants and was a limitation for both studies. To further confirm the relationship and to foster external validity, we conducted another study in a real life situation to explore how a common altruistic behavior, donation, affected warmth perception of the ambient environment.

Study 4 was a field experiment conducted to increase the external validity of previous work and confirm the effects of altruistic behavior on the perception of warmth of the ambient environment. The experiment was conducted with the help of the Students Association of Sunshine Volunteers at Peking University. An actual charity event was held on December 7, 2012, to collect donations for children with leukemia. A donation desk was placed on the sidewalk along the campus main street. We randomly solicited individuals to complete a short questionnaire after they passed the donation desk. We hypothesized that people who donated (the altruistic group) would feel warmer regarding the ambient environment than those who did not donate (the non-altruistic group).

Because Study 4 was a field study, a power of 0.8 and an effect size d of 0.5 (which was smaller than previous studies) were used to compute the expected sample size, resulting in a sample size of 102. A total of 108 people participated in this experiment (47 males; M a ge = 22.27 years, SD = 5.50). Of these, 55 made a donation.

Data were collected from 9:00 a.m. to 1:00 p.m. during the day. The environment temperature during the experiment ranged from -7°C to 0°C. Participants filled in a short questionnaire called “Perception of Weather Conditions in Beijing,” which included items assessing warm feelings about the environment along an 11-point scale (0 = extremely cold , 10 = extremely warm) and an estimation of the environment temperature in Celsius degrees. To control for hourly temperature variations, we paired a passer-by who did not donate within 2 min after a donor completed the questionnaire.

Consistent with our prediction, results showed that the passers-by who made a donation (altruists) perceived the ambient environment as warmer ( M = 3.07, SD = 2.01) than those who did not ( M = 2.40, SD = 1.86). The difference was marginally significant, t (106) = 1.81, p = 0.073, d = 0.35. The donors also reported significantly higher temperature estimations ( M = -0.51°C, SD = 3.40) compared to the non-donors ( M = -2.34°C, SD = 3.33), t (106) = 2.83, p = .006, d = 0.54).

A Meta-Analysis

So far, we have consistently revealed the effects of altruistic behavior on physical warmth. For further verification, we conducted a meta-analysis to test the statistical replication of the experiments. The present experiments provided a good condition for such an analysis for two reasons. First, the altruistic behaviors varied in category, including sharing food with others, charitable helping, and monetary donations. Second, both college students in China and residents in America were included in the experiments, offering a significant diversity of participant populations.

Because Study 2a did not involve grouping participants, data from all other five studies were included. We used warmth perception of the ambient environment as the main dependent variable. Using Comprehensive Meta Analysis software, we entered the means, standard deviations, and sample sizes of the altruistic group and the non-altruistic group to calculate effect sizes. As shown in Table 1 , the combined z -value was 5.94 ( p < 0.001), which confirmed that altruistic behaviors increased physical warmth. Moreover, the heterogeneity test showed that the q -value was 6.24 ( p = 0.182), indicating that the effects did not differ significantly across the studies.

TABLE 1. Results of a mini meta-analysis of the six studies.

General Discussion

Through six experiments, we found that altruistic performers were likely to feel warmer about the ambient environment than those who either refused to help or did not have a chance to help. Furthermore, results revealed that this effect was mediated by perceived social distance (Studies 2b and 3a). Moreover, this effect was confirmed in both crisis and ordinary situations amidst different experimental settings. Altruism is a complicated construct. It is usually directed by voluntary motivations and thus altruistic behaviors were just observed in Studies 2a, 3a, and 4. To confirm the causal relationship, recalled altruistic experiences as well as instant altruistic behaviors were manipulated in Studies 1, 2b, and 3b. Moreover, different instructions and cover stories were incorporated in Studies 2b and 3b. In summary, this design of series studies complemented for possible weaknesses in terms of constructing altruism and offered convergent findings.

Immediate Self-Reward of Altruistic Behaviors

While improving the welfare of others, altruistic behaviors typically deplete the energy and resources of the altruistic performers. Earlier explanations of altruism focused on its long-term return, either through genetic propagation ( Hamilton, 1963 ; Hamilton and Axelrod, 1981 ) or reciprocity ( Trivers, 1971 ; Nowak and Sigmund, 1998 , 2000 ). However, research on altruistic punishment observed the activation of the reward system of the dorsal striatum, signaling an instant satisfaction from altruistic behaviors ( de Quervain et al., 2004 ). Results from the current research have implied a potential self-reward mechanism for altruistic behaviors.

We found an immediate rewarding effect on the perceived warmth of the ambient environment after performing altruistic behaviors. In six studies, warm feelings were measured immediately after the participants reported the willingness to help and after they exhibited actual altruistic behaviors. This suggested that the increased feeling of warmth surrounding the ambient environment was an immediate reward rather than a long-term return-benefit for altruistic behavior performers. Furthermore, this effect was found to be mediated by perceived social distance, introducing a potential psychological process in this immediate reward. As mentioned in the introduction, previous research also found that altruistic behavior could result in some immediate psychological reward, including happiness ( Dunn et al., 2008 ), self-efficacy ( Midlarsky and Kahana, 1994 ), and positive self-evaluation ( Post, 2005 ). Moreover, Dawans et al. (2012) found that exposure to acute social stress could increase prosocial responses, which implied an immediate protective function of altruism in coping with acute stress. In the current research, the immediate reward was found to extend to concrete physical feelings through a psychological process.

We proposed three characteristics of the immediate reward system of altruism that were different from those of its long-term benefits. First, an immediate reward is a much more certain and spontaneous effect compared with long-term benefits. An altruistic performer may end up with no real benefits in the long run but could always obtain positive feedbacks through the immediate reward system. Second, the immediate reward could operate on the psychological or even perceptual level. Compared with the external benefits from genetic propagation and reciprocity, the internal immediate reward could serve as a direct incentive to engage in altruistic behaviors. Third, receiving an immediate reward could be a self-feedback process. For long-term benefits, altruistic performers usually need to rely on the behaviors or the survival possibilities of others. In contrast, altruistic behaviors could directly activate the performer’s psychological and physical processes on the individual level. To summarize, an immediate reward from altruism could be certain, internally perceived, and self-activated, compared to long-term benefits.

The immediate reward for altruistic performers could be very valuable. The observed results of altruistic behaviors in previous research included positive emotions and positive self-cognitions, all of which were found to be positively related with people’s subjective well-being (e.g., Bandura, 1986 ; Brunstein, 1993 ). A meta-analysis found that volunteers had lower risks of mortality ( Jenkinson et al., 2013 ). In the current research, we observed a consistent and direct connection between physical warmth and altruism. Physical warmth may be an important source of psychological energy or at least a strong comfort for individuals who were exposed to a cold environment. Specifically, individuals could perceive the environment as more secure and predictable ( IJzerman et al., 2015a ), and thus gain more confidence in coping with the environment. Thus, if the immediate reward from altruism could be valuable for the altruistic performers, they can be trusted because the reward would be more direct and certain.

In all, it suggests that altruism can not only bring long-term benefits to the performers (as explained in kin selection and reciprocity accounts) but also evoke immediate positive feelings inside the performers (as revealed in the current research). These two systems offer a better understanding of the functional adaptabilities of altruistic behaviors. Traditional economic analysis of altruism is based on a cost/utility analysis of external factors such as money, time, and probabilities of genetic propagations. We proposed to add internal utility (psychological states such as warmth perception and emotion) in the discussion of altruism. Altruistic behaviors would be evolutionarily meaningful if the increase in the internal utility offsets the reductions in external utility. Thus, the framework of altruism would be more complete when we concern both long-term and immediate benefits, as well as both external and internal utilities.

Lingering Fragrance Effect

A potential immediate self-reward system of altruistic behaviors has practical meaning in promoting people’s well-being and quality of life in both daily life and times of crisis. The effect found in the current research was a typical example of the interaction between an individual’s physical and psychological system. The saying the roses in her hand, the flavor in mine could reveal a real psychological activation effect, through which people could promote their personal physical states. Thus, we named this psychological activation effect as a lingering fragrance effect.

When facing a threat (e.g., hunger, pain, or cold), there are mainly two strategies for its removal. One is to cope with the threat immediately (e.g., put on a coat to remove coldness). The other strategy is to activate a psychological process to change the perception of the threat. As in the current research, altruistic behavior activated people’s psychological processes and changed their warmth feelings of a cold environment. During crisis, the adverse external conditions could hardly be changed because of insufficient support of food, water, and living conditions. However, individual cognitions and behaviors are pliant and controllable, offering possibilities to attenuate the threats in crisis. Hence, psychological activations or subjective adjustments could become a prospective coping strategy under such circumstances. Importantly, the experimental settings used to induce people’s real altruistic behaviors might be artificial in the current research. Moreover, we found consistent lingering fragrance effects in spite of this shortcoming. Thus we expect in real life that the lingering fragrance effect of altruistic behaviors should be amplified to benefit individuals in a more extensive manner.

To note, we focused on the effects of altruistic behaviors on individuals’ subjective feelings of warmth in the current study. And we did not found significant similar effects on individuals’ objective physical states (body temperature in Studies 3a and 3b). This also implied that the psychological reactions may be faster and more flexible than physical reactions. However, it is possible that the physical states could get feedbacks from psychological states when the timeline is extended concerning the close interactions of physical and psychological states (e.g., Edwards and Cooper, 1988 ) and the prospective treatment of biofeedback (e.g., Lagos et al., 2013 ). In addition, we have mentioned in the introduction as well as in the discussion section that an increase in warmth feelings could be helpful for individuals who were coping with cold environment. Warmth feelings could serve as an easily accessible comfort for them especially in some crisis situations. On the other hand, it could become a problem when individuals optimistically perceived the environment as warm but failed to respond to the important environmental cue about temperature. This indicates that a raise of warmth feelings may actually become a double-edge sword in some cases. And it is interesting and prospective to explore the other edge of the sword in future studies.

In summary, the lingering fragrance effect is an important perspective for understanding the strategy in coping with a threat. The interaction between physical and psychological systems makes it possible to change the physical state through the mobilization of internal psychological resources. Under the crisis circumstances with limited conditions, using a psychological resource to resist the bad impacts of a crisis could be a more reasonable or even a single possible strategy to cope with threats. As the results showed, active altruistic behavior is a significant way for people to resist the cold environment. According to the Harry Truman effect, a person’s potential can be activated under certain circumstance ( Seligman, 2002 ). Equally, a person can activate his/her own internal psychological potential to cope with the intense threat and pursue positive results.

Future Directions

The exploration of the proposed immediate self-reward model of altruistic behaviors could be expanded from two aspects. First, the crisis situations and the experimental settings in the current research were all related to coldness. In some cases, coldness could even be a significant threat to survival (e.g., Hurricane Sandy in Study 1 and the earthquake in Studies 2a and 2b). In such cases, increased feelings of ambient warmth were regarded as a reward for the individuals. The specific reward and the corresponding psychological processes might change with different adverse situations. For example, coolness could become a more comfortable state when the environment was extremely hot. Future research could help to explore different contents of the lingering fragrance effect and to offer more evidence for an immediate internal reward system.

Second, further studies could be conducted on the rewarding nature of altruism. For example, a positive cycle might be established in which increased warmth perception following altruistic behaviors result in future altruistic behaviors. Moreover, de Quervain et al. (2004) found that effective altruistic punishment was related to stronger activation of the dorsal striatum. Similarly, neuroscientific studies would be helpful in finding out whether performing an altruistic behavior would activate the same region or some other reward-related brain regions.

Ethics Statement

The research project was approved by the Ethics Committee of Department of Psychology, Peking University. Studies 3a, 3b, and 4 involved helping disadvantaged groups (i.e., children from low-income migrant workers’ families, students from a remote and poor town, and children with leukemia). All of the helping scenarios used in the studies (including refining education materials, sending postcards, and donations) were actually carried out.

Author Contributions

T-YH, JL, HJ, and XX should be considered as joint first authors as they contributed equally to the research.

This work was supported by the Key Program of National Natural Science Foundation of China [grant number 91224002]; and the General Program of National Nature Science Foundation of China [grant number 71172024].

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

We would like to appreciate Xiaoxiao Hu for helping us with data collection. Also, many thanks to Student Union of Psychology in Peking University, Jimunai Middle School and Qizhi Ning, who offered us opportunities to conduct Studies 3a, 3b, and 4 and to help those kids who were in need. Last but not least, we would like to extend our sincere gratitude to the members of Risk and Decision-making Lab of Psychology Department of Peking University for their helpful comments and support on the research.

• ^ http://www.mturk.com
• ^ Participants in Study 2b took the experiment in group of 4 (three participants and one confederate) and they needed to act to share food during the experiment. Although participants were asked not to interact with others, it was possible that their reactions were influenced by other participants. To rule out this effect, we run an HLM test with a contextual model. The average number of packs within each group was regarded as the contextual variable. Results showed that the effects of altruism manipulation was marginally significant, t (77) = 1.84, p = 0.069. The effect of contextual variable did not reached significance, t (77) = 0.18, p = 0.860. The results remained when negative emotions, perceived severity and uncertainty of the scenario were included in the model.

Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall.

Google Scholar

Bargh, J. A., and Shalev, I. (2012). The substitutability of physical and social warmth in daily life. Emotion 12, 154–162. doi: 10.1037/a0023527

CrossRef Full Text | Google Scholar

Barsalou, L. W. (1999). Perceptual symbol system. Behav. Brain Sci. 22, 577–609. doi: 10.1017/S0140525X99532147

Barsalou, L. W. (2008). Cognitive and neural contributions to understanding the conceptual system. Curr. Dir. Psychol. Sci. 17, 91–95. doi: 10.1111/j.1467-8721.2008.00555.x

Basbaum, A. I., and Fields, H. L. (1984). Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu. Rev. Neurosci. 7, 309–338. doi: 10.1146/annurev.ne.07.030184.001521

Batson, C. D. (1991). The Altruism Question: Toward a Social-Psychological Answer. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Beckes, L., and Coan, J. A. (2011). Social baseline theory: the role of social proximity in emotion and economy of action. Soc. Pers. Psychol. Compass 5, 976–988. doi: 10.1111/j.1751-9004.2011.00400.x

Boyd, R., Gintis, H., Bowles, S., and Richerson, P. J. (2003). The evolution of altruistic punishment. Proc. Natl. Acad. Sci. U.S.A. 100, 3531–3535. doi: 10.1073/pnas.0630443100

Brunstein, J. C. (1993). Personal goals and subjective well-being: a longitudinal study. J. Pers. Soc. Psychol. 65, 1061–1070. doi: 10.1037/0022-3514.65.5.1061

Chamarro, A., and Fernández-Castro, J. (2009). The perception of causes of accidents in mountain sports: a study based on the experiences of victims. Accid. Anal. Prev. 41, 197–201. doi: 10.1016/j.aap.2008.10.012

Cohen, J. (1992). A power primer. Psychol. Bull. 112, 155–159. doi: 10.1037/0033-2909.112.1.155

Danielli, J. F. (1980). Altruism and the internal reward system or the opium of the people. J. Soc. Biol. Struct. 3, 87–94. doi: 10.1016/0140-1750(80)90001-9

Dawans, B., Fischbacher, U., Kirschbaum, C., Fehr, E., and Heinrichs, M. (2012). The social dimension of stress reactivity: acute stress increases prosocial behavior in humans. Psychol. Sci. 23, 651–660. doi: 10.1177/0956797611431576

de Quervain, D. J. F., Fischbacher, U., Treyer, V., Schellhammer, M., Schnyder, U., Buck, A., et al. (2004). The neural basis of altruistic punishment. Science 305, 1254–1258. doi: 10.1126/science.1100735

de Waal, F. B. M. (2008). Putting the altruism back into altruism: the evolution of empathy. Annu. Rev. Psychol. 59, 279–300. doi: 10.1146/annurev.psych.59.103006.093625

Delgado, M. R., Locke, H. M., Stenger, V. A., and Fiez, J. A. (2003). Dorsal striatum responses to reward and punishment: effects of valence and magnitude manipulations. Cogn. Affect. Behav. Neurosci. 3, 27–38. doi: 10.3758/CABN.3.1.27

Dunn, E. W., Aknin, L. B., and Norton, M. I. (2008). Spending money on others promotes happiness. Science 319, 1687–1688. doi: 10.1126/science.1150952

Edwards, J. R., and Cooper, C. L. (1988). The impact of positive psychological states on physical health: a review and theoretical framework. Soc. Sci. Med. 27, 1447–1459. doi: 10.1016/0277-9536(88)90212-2

Gintis, H. (2000). Strong reciprocity and human sociality. J. Theor. Biol. 206, 169–179. doi: 10.1006/jtbi.2000.2111

Gintis, H., Bowles, S., Boyd, R., and Fehr, E. (2003). Explaining altruistic behavior in humans. Evol. Hum. Behav. 24, 153–172. doi: 10.1016/S1090-5138(02)00157-5

Hamilton, W. D. (1963). The evolution of altruistic behavior. Am. Nat. 97, 354–356. doi: 10.2307/2458473

Hamilton, W. D., and Axelrod, R. (1981). The evolution of cooperation. Science 211, 1390–1396. doi: 10.1126/science.7466396

Harbaugh, W. T. (1998). What do donations buy? A model of philanthropy based on prestige and warm glow. J. Pubic Econ. 67, 269–284. doi: 10.1016/S0047-2727(97)00062-5

Harlow, H. F. (1958). The nature of love. Am. Psychol. 13, 673–685. doi: 10.1037/h0047884

Heir, H., Piatigorsky, A., and Weisæth, L. (2009). Longitudinal changes in recalled perceived life threat after a natural disaster. Br. J. Psychiatry 194, 510–514. doi: 10.1192/bjp.bp.108.056580

Henrich, J., McElreath, R., Barr, A., Ensminger, J., Barrett, C., Bolyanatz, A., et al. (2006). Costly punishment across human societies. Science 312, 1767–1770. doi: 10.1126/science.1127333

IJzerman, H., Coan, J. A., Wagemans, F. M., Missler, M. A., van Beest, I., Lindenberg, S., et al. (2015a). A theory of social thermoregulation in human primates. Front. Psychol. 6:464. doi: 10.3389/fpsyg.2015.00464

Ijzerman, H., Gallucci, M., Pouw, W. T., Weibetagerber, S. C., Van Doesum, N. J., and Williams, K. D. (2012). Cold-blooded loneliness: social exclusion leads to lower skin temperatures. Acta Psychol. 140, 283–288. doi: 10.1016/j.actpsy.2012.05.002

IJzerman, H., Janssen, J. A., and Coan, J. A. (2015b). Maintaining warm, trusting relationships with brands: increased temperature perceptions after thinking of communal brands. PLoS ONE 10:e0125194. doi: 10.1371/journal.pone.0125194

IJzerman, H., and Semin, G. R. (2009). The thermometer of social relations: mapping social proximity on temperature. Psychol. Sci. 20, 1214–1220. doi: 10.1111/j.1467-9280.2009.02434.x

IJzerman, H., and Semin, G. R. (2010). Temperature perceptions as a ground for social proximity. J. Exp. Soc. Psychol. 46, 867–873. doi: 10.1016/j.jesp.2010.07.015

Inagaki, T. K., and Eisenberger, N. I. (2013). Shared neural mechanisms underlying social warmth and physical warmth. Psychol. Sci. 24, 2272–2280. doi: 10.1177/0956797613492773

Jenkinson, C. E., Dickens, A. P., Jones, K., Thompson-Coon, J., Taylor, R. S., Rogers, M., et al. (2013). Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers. BMC Public Health 13:773. doi: 10.1186/1471-2458-13-773

Kang, Y., Williams, L. E., Clark, M. S., Gray, J. R., and Bargh, J. R. (2011). Physical temperature effects on trust behavior: the role of insula. Soc. Cogn. Affect. Neurosci. 6, 507–515. doi: 10.1093/scan/nsq077

Kasperson, R. E., Renn, O., Slovic, P., Brown, H. S., Emel, J., Goble, R., et al. (1988). The social amplification of risk: a conceptual framework. Risk Anal. 8, 177–187. doi: 10.1111/j.1539-6924.1988.tb01168.x

Kellogg, B. (2013). Sandy’s Snowy Side Turns Deadly. Available at: http://www.weather.com/storms/hurricane/news/sandy-snowy-side-20121030

Knutson, B., Westdorp, A., Kaiser, E., and Hommer, D. (2000). FMRI visualization of brain activity during a monetary incentive delay task. Neuroimage 12, 20–27. doi: 10.1006/nimg.2000.0593

Lagos, L., Thompson, J., and Vaschillo, E. (2013). A preliminary study: heart rate variability biofeedback for treatment of postconcussion syndrome. Biofeedback 41, 136–143. doi: 10.5298/1081-5937-41.3.02

Landau, M. K., Meier, B. P., and Keefer, L. A. (2010). A metaphor-enriched social cognition. Psychol. Bull. 136, 1045–1067. doi: 10.1037/a0020970

Leknes, S., and Tracey, I. (2008). A common neurobiology for pain and pleasure. Nat. Rev. Neurosci. 9, 314–320. doi: 10.1038/nrn2333

MacKinnon, D. P., Fairchild, A. J., and Fritz, M. S. (2007). Mediation analysis. Annu. Rev. Psychol. 58, 593–614. doi: 10.1146/annurev.psych.58.110405.085542

Midlarsky, E., and Kahana, E. (1994). Altruism in Later Life. Thousand Oaks, CA: Sage.

Molin, J., Mellerup, E., Bolwig, T., Scheike, T., and Dam, H. (1996). The influence of climate on development of winter depression. J. Affect. Disord. 37, 151–155. doi: 10.1016/0165-0327(95)00090-9

Myers, D. (1993). Social Psychology , 4th Edn. New York, NY: McGraw-Hill.

Nowak, M. A., and Sigmund, K. (1998). Evolution of indirect reciprocity by image scoring. Nature 393, 573–577. doi: 10.1038/31225

Nowak, M. A., and Sigmund, K. (2000). Shrewd investments. Science 288, 819–820. doi: 10.1126/science.288.5467.819

Post, S. G. (2005). Altruism, happiness, and health: it’s good to be good. Int. J. Behav. Med. 12, 66–77. doi: 10.1207/s15327558ijbm1202_4

Schubert, T. W. (2005). Your highness: vertical positions as perceptual symbols of power. J. Pers. Soc. Psychol. 89, 1–21. doi: 10.1037/0022-3514.89.1.1

Seligman, M. E. (2002). Authentic Happiness: Using the New Positive Psychology to Realize Your Potential for Lasting Fulfillment. New York, NY: Simon and Schuster.

Szymkow, A., Chandler, J., IJzerman, H., Parzuchowski, M., and Wojciszke, B. (2013). Warmer hearts, warmer rooms: how positive communal traits increase estimates of ambient temperature. Soc. Psychol. 44, 167–176. doi: 10.1027/1864-9335/a000147

Trivers, R. L. (1971). The evolution of reciprocal altruism. Q. Rev. Biol. 46, 35–57. doi: 10.2307/2822435

Van Lange, P. A. M., Otten, W., De Bruin, E. M. N., and Joireman, J. A. (1997). Development of prosocial, individualistic, and competitive orientations: theory and preliminary evidence. J. Pers. Soc. Psychol. 73, 733–746. doi: 10.1037/0022-3514.73.4.733

Ward, V. (2015). Winter Death Toll ‘to Exceed 40,000’. The Telegraph. Available at: http://www.telegraph.co.uk/news/weather/11382808/Winter-death-toll-to-exceed-40000.html

Webb, R. (2014). Death Toll Rises from Harsh Winter Weather. USA Today. Available at: http://www.usatoday.com/story/weather/2014/01/07/cold-storms-deaths/4355329

Williams, L. E., and Bargh, J. A. (2008). Experiencing physical warmth promotes interpersonal warmth. Science 322, 606–607. doi: 10.1126/science.1162548

Zhong, C. B., and Leonardelli, G. J. (2008). Cold and lonely: does social exclusion literally feel cold? Psychol. Sci. 19, 838–842. doi: 10.1111/j.1467-9280.2008.02165.x

Keywords : altruistic behavior, altruistic performer, crisis, social distance, warmth perception

Citation: Hu T-Y, Li J, Jia H and Xie X (2016) Helping Others, Warming Yourself: Altruistic Behaviors Increase Warmth Feelings of the Ambient Environment. Front. Psychol. 7:1349. doi: 10.3389/fpsyg.2016.01349

Received: 21 February 2016; Accepted: 23 August 2016; Published: 05 September 2016.

Reviewed by:

Copyright © 2016 Hu, Li, Jia and Xie. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Xiaofei Xie, [email protected].

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Norm enforcement in the city revisited: An international field experiment of altruistic punishment, norm maintenance, and broken windows

Berger, Joël ; Hevenstone, Debra (2016). Norm enforcement in the city revisited: An international field experiment of altruistic punishment, norm maintenance, and broken windows. Rationality and society, 28(3), pp. 1-21. Sage Publications 10.1177/1043463116634035

This is the latest version of this item.

In laboratory experiments, people are willing to sanction norms at a cost—a behavioral tendency called altruistic punishment. However, the degree to which these findings can be generalized to real-world interactions is still debated. Only a small number of field experiments have been conducted, and initial results suggest that punishment is less frequent outside of the lab. This study replicates one of the first field experiments on altruistic punishment and builds ties to research on norm compliance and the broken windows theory. The original study addressed the enforcement of the anti-littering norm in Athens. We replicate this study in Bern, Zurich, and New York City. As an extension, we investigate how the experimental context (clean vs littered) impacts social norm enforcement. As a second extension, we investigate how opportunity structure impacts the maintenance of the anti-littering norm. Findings indicate that norms are universally enforced, although significantly less than in the standard laboratory experiment,and that enforcement is significantly more common in Switzerland than in New York. Moreover, individuals prefer more subtle forms of enforcement to direct punishment. We also find that enforcement is less frequent in littered than in clean contexts, suggesting that broken windows might not only foster deviant behavior but also weaken informal social control. Finally, we find that opportunity structure can encourage people to maintain norms, as indicated by the fact that people are more likely to voluntarily pick up litter when it is closer to a trash bin.

Interest & Impact

5 since deposited on 11 May 2016 0 in the past 12 months

5 Citations in Web of Science ® 6 Citations in Scopus

in Google Scholar™

Actions (login required)

Available Versions of this Item

• Norm enforcement in the city revisited: An international field experiment of altruistic punishment, norm maintenance, and broken windows. (deposited 11 May 2016 15:47) [Currently Displayed]

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Rewarding altruism? A natural field experiment

ABSTRACT We present evidence from a natural field experiment involving nearly 100,000 individuals on the effects of offering economic incentives for blood donations. Subjects who were offered economic rewards to donate blood were more likely to donate, and more so the higher the value of the rewards.

Related Papers

Juan Cabasés

Fariba Binte Kawsar

Experimental studies document that financial rewards discourage the performance of altruistic activities, because they destroy intrinsic altruistic motivations. We set up a randomized-controlled experiment, through a survey administered to 467 blood donors in an Italian town, and find that donors are not reluctant to receive compensation in general: A substantial share of respondents declared they would stop being donors if paid a small amount of cash, but we do not find such effects when a voucher of the same nominal value is offered instead. The aversion to direct cash payments is particularly marked among women and older respondents, while there are neither gender nor age differences in the response to the voucher. Implications for research and public policy are discussed.

The Journal of Socio-Economics

Friedel Bolle , Philipp E Otto

Nicola Lacetera

This paper analyzes the e¤ects of a legislative provision that grants a one-day paid leave of absence to blood donors who are employees in Italy, using a unique dataset with the complete donation histories of the blood donors in an Italian town. The across-donor variation in job market status, and within-donor job switching over time are the sources of variation that we employ to study whether the paid-day-o ¤ incentive a¤ects the frequency of their donations. Our analysis indicates that the day-o ¤ privilege leads donors who are employees to make, on average, one extra donation per year, which represents an increase of around 40%. We discuss the implications of our ndings for policies aimed at reducing the shortages in the supply of blood and, more generally, for incentivizing pro-social behavior.

Mireia Jofre-Bonet

Transfusion Medicine and Hemotherapy

Daniel Farrelly

Research shows that altruistic behaviours arise in varying social situations in line with different theories of causes of such behaviours. However most research uses financial costs only, which makes our understanding of altruism currently limited. This study presents findings of three experiments that use a novel and simple laboratory-based task that measures altruism based on the amount of time participants are willing to spend as a cost to help others. This task assessed two specific theories; altruistic punishment (Experiments 1 & 2) and empathy-altruism (Experiment 3). All experiments showed that the task was successful, as participants were more likely to altruistically punish violators of social contracts than other scenarios (Experiments 1 and 2), and also incur more costs to behave altruistically towards others when feeling empathic than different emotional states (Experiment 3). These results provide clear support for the use and value of this novel task in future research.

Transfusion medicine and hemotherapy : offizielles Organ der Deutschen Gesellschaft fur Transfusionsmedizin und Immunhamatologie

Robert L. Mayo

I conduct an on-line experiment to decompose giving in a dictator game into amounts motivated by pure altruism and amounts motivated by the warm glow of giving as defined in Andreoni (1989). By manipulating the price of benefit to the recipient while holding the price of the act of giving constant, I estimate separate values for giving motivated by warm glow and giving motivated by pure altruism,. I find significant evidence of both pure altruism and warm glow as motivations for the amounts sent to an anonymous recipient. However, I also find a large gender difference in the motivation for giving. Females are significantly less sensitive to the price of benefit to the recipient than are males, suggesting females are motivated relatively more by warm glow and relatively less by pure altruism, while men display the opposite behavior.

RELATED PAPERS

Economics Bulletin

Katerina Sherstyuk , Malcolm Dowling

Southern Economic Journal

S. Ball , Richard Ashley , Catherine Eckel

James Andreoni

Axel Mühlbacher

Sao-Wen Cheng

Robert Slonim

SSRN Electronic Journal

Glenn Harrison

Nursing Ethics

Navaz Naghavi

Journal of Theoretical Biology

Miguel Fuentes Cortés

Econometrica

John Miller

Wim de Kort

International Journal of Research in Marketing

Michel Clement

Organizational Behavior and Human Decision Processes

Ayelet Gneezy

Erik O Kimbrough

Theory and Decision

M. Vlassopoulos

Ted Bergstrom

Vox Sanguinis

Vit Rehacek

John Elmer Lacson

Mark Ottoni-Wilhelm

Experimental Economics

Matthias Benz

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

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
• Front Psychol

Helping Others, Warming Yourself: Altruistic Behaviors Increase Warmth Feelings of the Ambient Environment

1 School of Psychological and Cognitive Sciences and Beijing Key Laboratory of Behavior and Mental Health, Peking University, Beijing, China

2 College of Education, Shanghai Normal University, Shanghai, China

Huiyuan Jia

3 School of Business Administration, Capital University of Economics and Business, Beijing, China

Xiaofei Xie

Altruistic behaviors typically improve the welfare of the recipient at the cost of the performer’s resources and energy. Do altruistic performers obtain any positive internal reward from altruistic behaviors? We conducted six experiments to explore whether altruistic behaviors could increase performer’s warmth perception of the ambient environment. The first three studies focused on crisis situations. A retrospective field study (Study 1, with Hurricane Sandy) and two laboratory studies (Studies 2a and 2b, with an earthquake scenario) found that people who helped others felt warmer of the ambient environment than people who did not. We extended to daily life situations and found that participants who performed helping behaviors in laboratory (either voluntarily in Study 3a or randomly assigned to in Study 3b) and passers-by who donated to a charity (Study 4) reported warmer perception of the ambient environment than those who did not. These findings suggested an immediate internal reward of altruism.

Introduction

One who watereth will himself be watered. Proverbs 11:25

When we talk about altruistic behaviors, we often talk about sacrifice and the potential costs and risks associated with it. At the individual level, altruism is often non-economic and even maladaptive to survival because the altruistic performer needs to share his/her own resources and energy with others without receiving explicit returns ( Batson, 1991 ; Myers, 1993 ; de Waal, 2008 ). When disasters signal a shortage of resources, it seems unwise to behave altruistically. However, at many times, we quote the saying the roses in her hand, the flavor in mine to encourage more altruistic behaviors. Is this quote simply a fortune cookie comment? Could those who behave altruistically literally experience the “flavor” or other positive physical feelings, as implied in this quote? In the current research, we explored the effects of altruistic behaviors on individuals’ physical feelings, focusing in particular on warmth. Warmth is a fundamental need to humans and other primates (e.g., Harlow, 1958 ; IJzerman et al., 2015a ). Moreover, warmth perception of the ambient environment is a typical variable that links the psychological and physical worlds. Specifically, we found that altruistic behaviors could lead the altruistic performer to increase his/her warmth perception of the ambient environment.

The Functional Adaptabilities of Altruistic Behaviors

Researchers have tried to recognize the functional adaptabilities of altruistic behaviors despite the self-sacrificing nature of such behaviors. The kin selection theory ( Hamilton, 1963 ; Hamilton and Axelrod, 1981 ) suggests that altruistic behaviors toward those with shared genes (i.e., offspring or relatives of the altruistic performer) could maximize genetic frequencies at the group level. The reciprocal motivation and social exchange theory suggests that an altruistic performer could expect future returns either directly from the recipient ( Trivers, 1971 ) or indirectly from a third party ( Nowak and Sigmund, 1998 , 2000 ). The above two theories both focus on the long-term benefits for the altruistic performers ( Gintis et al., 2003 ).

However, researchers have uncovered the phenomenon of altruistic punishment. Individuals chose to incur great cost to punish norm violators in a group (e.g., Henrich et al., 2006 ). Such choices were regarded as altruistic for they aimed at restoring fairness and protecting the group norm. Because there was usually no kin relative in the group and the research employed a one-shot game ( Gintis, 2000 ; Boyd et al., 2003 ; Gintis et al., 2003 ; Henrich et al., 2006 ), this phenomenon could not be explained by the two previously mentioned theories. Moreover, a neuroscientific study on altruistic punishment found that effective punishment activated the dorsal striatum. Participants with stronger activation in this brain area were willing to perform more punishment ( de Quervain et al., 2004 ). As the dorsal striatum is an important part of the reward system ( Knutson et al., 2000 ; Delgado et al., 2003 ), the findings implied that altruistic punishment could trigger immediate positive experiences for altruistic performers. This finding was consistent with earlier notions that altruistic behaviors would promote the release of endogenous opioid peptides ( Danielli, 1980 ), which contributed to the control of pain ( Basbaum and Fields, 1984 ) and the modulation of human mood and feelings of well-being ( Leknes and Tracey, 2008 ). Furthermore, these findings implied that we may focus on the internal rewarding system of human altruistic behaviors. Rather than uncover tangible returns as a result of altruistic behaviors, we would like to explore the potential positive effects of altruism on one’s psychological and physical experiences.

Researchers have found positive psychological consequences of altruistic behaviors. For example, prosocial spending including charity donations and gift giving were found to evoke happiness ( Dunn et al., 2008 ). Researchers in this field used the term “warm glow” to indicate an internal sense of satisfaction in donors after donating money ( Harbaugh, 1998 ). Altruistic behaviors also promoted self-efficacy in the elderly ( Midlarsky and Kahana, 1994 ) and enhanced positive self-evaluations ( Post, 2005 ). In the current study, we focused on the warmth perception of the ambient environment. We chose this variable for three reasons. First, warmth is a fundamental need of humans and other primates. This has been supported by earlier studies on development and attachment (e.g., Harlow, 1958 ) as well as by a recent model of thermoregulation ( IJzerman et al., 2015a ). Feeling of warmth could be a source of security ( Harlow, 1958 ) and individuals could use feeling of warmth as an indicator of social resources ( IJzerman et al., 2015a ). Research on winter depression implied that lack of warmth could be a threat to mental health (e.g., Molin et al., 1996 ). If altruism were to lead to threats against other survival-related resources such as food, compensatory feelings of warmth could be a comfort for individuals who are facing crisis situations, which could be an advantage for survival. Second, it is implied from previous research that the reward of altruism might be complicated. Reward, for instance, might be related to an internal reward system rather than simply tangible resources. Warmth perception of the ambient environment is a variable that links the psychological and physical worlds. It describes individuals’ internal reflections of the surrounding world, which could be an important facet of positive feedback of altruism. Third, coldness is a typical threat in many crises, especially in natural environments. For example, extreme cold weather in winter could cause a large amount of deaths (e.g., Webb, 2014 ; Ward, 2015 ). Coldness is also one of the most important environmental factors in mountain sports accidents ( Chamarro and Fernández-Castro, 2009 ). Compared with non-crisis situations, people are more likely to lack resources (e.g., warm food, warm clothes) in coping with coldness. Moreover, individuals may tend to amplify the threats of the crisis-situations. Such tendency was usually related to other psychological reactions toward disasters such as post-traumatic stress disorder (e.g., Heir et al., 2009 ). And the tendency could also be amplified in a social level ( Kasperson et al., 1988 ). Under extreme circumstances, individuals can do little to change the objective situation. Regulating warmth feelings of the ambient environment is potentially helpful for individuals to feel positively about the situations. We hypothesized that altruistic behaviors would increase the feelings of warmth among altruistic performers.

Altruistic Behaviors Promote Physical Warmth

Researchers have linked feelings of warmth to social behaviors and social cognitions. Recently, IJzerman et al. (2015a) proposed a social thermoregulation model. According to this model, thermoregulation is costly for a single individual; therefore, social interactions (e.g., bodily contact) are vital and economic for animals to maintain proper body temperatures. The model argues that the process of social thermoregulation has shaped high-order social cognition. Earlier theories on grounded cognition ( Barsalou, 1999 , 2008 ; Schubert, 2005 ) have suggested that abstract social cognitions are grounded in interactions with the physical world. Metaphoric models emphasize metaphoric mappings between social cognition and physical perception ( Landau et al., 2010 ). These theories help explain the connections between the social and physical worlds in human beings.

An important line of research has focused on a bidirectional relationship between physical warmth and social warmth. On the one hand, a well-known study conducted by Williams and Bargh (2008) showed that a simple manipulation of physical warmth could lead to favorable interpersonal evaluations and behaviors. In addition, Kang et al. (2011) found that physical warmth increased trusting behavior. Such behaviors were regarded conceptually as interpersonal warmth in social life. On the other hand, other studies have yielded evidence that social-related concepts raised warmth perception. For example, research found that thinking about traits and objects that were positively related to communal concepts raised perceived warmth of the ambient environment ( Szymkow et al., 2013 ; IJzerman et al., 2015b ). Moreover, neuroscientific studies have also supported connection between social warmth and physical warmth. Kang et al. (2011) found insula activation when physical warmth increased trusting behavior. Inagaki and Eisenberger (2013) ’s study also found that social-warmth and physical-warmth conditions overlapped on their activations in ventral striatum and middle insula. This overlap was specific in warmth-related positive feelings.

Furthermore, this social-physical link of warmth has been indirectly supported by the “cold” side of social life. Specifically, Zhong and Leonardelli (2008) found that recalling a past experience of social exclusion resulted in a lower estimate of ambient temperature. A social exclusion manipulation could lower skin temperature, and physical warmth experiences were more desirable and effective in comforting the participants experiencing feelings of loneliness ( Bargh and Shalev, 2012 ; Ijzerman et al., 2012 ).

Through the consistently observed relationship between social behaviors (especially positive or prosocial behaviors that foster social relationships) and feelings of warmth, we expected a similar relationship between altruistic behaviors (as typical prosocial behaviors) and warmth perception. Apart from happiness, self-efficacy or some other psychological states, altruistic behaviors could also lead to some positive consequences that are more physical related. Specifically, we predicted that altruistic behaviors would lead to an increased warmth perception of the ambient environment.

Perceived Social Distance as a Mediator

Apart from the research that confirmed a relationship between social warmth and physical warmth, many studies have identified the effects of social distance and physical distance on warmth feelings. For example, IJzerman and Semin (2010) found that a closer perception of social distance (either induced by physical distance or by similarity manipulation) resulted in a higher estimate of ambient temperature. The effects could be reasoned that there is a similar overlapping of social distance and physical interpersonal distance (e.g., IJzerman and Semin, 2009 ), and that physical warmth is usually related to physical interpersonal distance.

In a recent model, social baseline theory was built on previous theories of attachment and other related findings ( Beckes and Coan, 2011 ). The theory proposes that social relationship and social proximity is actually a baseline for human being because it is energy-saving and risk-reducing. Individuals are actually more activated when social relationships are threatened, and they have a tendency to return to “baseline.” Thus, individuals may incorporate relationship-oriented social behaviors to achieve social proximity. Because of the close link between social distance and warmth perception, social distance could be a proximal bridge between prosocial or relationship-oriented behaviors and warmth perception. Thus, we hypothesized that altruistic behaviors would increase the performers’ warmth perceptions of the ambient environment. Moreover, we predicted that this effect would be mediated by decreasing the perceived social distance between altruists and recipients.

Overview of the Current Research

In the current research, we conducted six studies to test the hypothesis that altruistic behaviors would result in a warmer perception of ambient environment and to additionally test the effect of perceived social distance as a mediator. Study 1 was a retrospective field study conducted in the context of Hurricane Sandy, revealing the relationship between altruistic behaviors and perceived warmth in a crisis. Then, in Studies 2a and 2b, we created a crisis situation in the laboratory to replicate this effect and to test the mediation effect of perceived social distance. In Studies 3a and 3b, we extended such effect and the mediation effect of perceived social distance to daily life situations using laboratory experiments. In Study 4, we conducted a field study to replicate such an effect in a real donation activity. To note, all the studies in the current research were set in a comparatively cold environment. We did not explore how this effect would manifest in a much hotter environment.

Study 1 explored the effect of altruistic behaviors on feelings of warmth in a real crisis context. Hurricane Sandy was the most destructive hurricane of the 2012 Atlantic hurricane season, dramatically shocking 24 states in the United States with its ferocity dating back to 29th October, 2012. After the storm crashed ashore, the rain turned into blizzard conditions along the east coast of the United States. According to news reports, the highest snowfall accumulation was 36 inches (91 cm) ( Kellogg, 2013 ). What was worse, due to the heavy snow, the heating and power systems were cut off for more than 10 days in some places in New York State. The extremely cold weather turned the recovery and reconstruction into chronic suffering. As a result, local residents were suffering the continuous stress of environmental threat in subsequent months. In this study, the sample included participants who were local residents at the time of the crisis and who experienced this crisis first-hand. We hypothesized that participants who recalled an altruistic experience (vs. those who recalled a non-altruistic experience) in the crisis would then have a memory of a warmer ambient environment.

Participants

G ∗ Power 3.1 was used to compute a priori power analyses in the study. According to Cohen’s (1992) suggestion, a power of 0.8 and an effect size that was slightly above moderate level (effect size d = 0.6) were used as the input data. The expected effect size was also consistent with previous research on similar topics (e.g., IJzerman and Semin, 2009 , 2010 ). This resulted in an expected sample size of 72. Seventy-nine local residents from the disaster areas (New Jersey State and New York State) (33 males; M a ge = 29.65 years, SD = 9.41) were recruited on Amazon’s Mechanical Turk website (MTurk 1 ). They were asked to complete an online survey named “Life after Sandy Hurricane” and were each rewarded a $5 Amazon gift-card. The purpose of the survey was described as for exploring how Hurricane Sandy and the following extreme cold weather influence people’s daily life. The survey was launched within 3 months after Hurricane Sandy hit the United States while residents in the disaster areas were still suffering through the chronic recovery and reconstruction. Participants were randomly assigned to the altruistic or non-altruistic group. There were 42 participants in the altruistic group and 37 participants in the non-altruistic group.

Materials and Procedure

Participants were first informed that the survey aimed to explore how Hurricane Sandy and the subsequent extreme weather conditions influenced the local residents. Next, participants were asked to recall an experience during or after Hurricane Sandy and to write it down in detail (using at least 50 words). In the altruistic group, participants were asked to “recall an experience in which he/she did something mainly taking other people’s benefits into consideration (e.g., giving food to a homeless person on the street, offering someone a free-ride, volunteer work, etc.).” In the non-altruistic group, they were asked to “recall an experience in which he/she did something mainly taking his/her own benefits into consideration (e.g., rushing to the supermarket for necessities without thinking about how others may need the goods, refusing to offer a free-ride, ignoring neighbor’s asking for help, etc.).” Afterward, participants were asked to recall their instant feelings of the ambient environment after the experience along a 7-point scale (1 = extremely cold , 7 = extremely warm ). As control variables, the participants were also asked to report their feelings of warmth in the current environment along a 7-point scale (1 = extremely cold , 7 = extremely warm ) in addition to the estimated temperature (in degrees Fahrenheit) of the current environment.

Results and Discussion

The participants recalled warmer ambient environment after the recalled altruistic experience ( M = 4.55, SD = 1.19) than after the recalled non-altruistic experience ( M = 3.35, SD = 1.06), t (77) = 4.68, p < 0.001, d = 1.06. The effect remained significant when we included as covariates the warmth feelings and the estimated temperature of the current environment during the survey, F (1,74) = 19.93, p < 0.001, η 2 = 0.21. Moreover, the warmth feelings or the estimated temperature of the current environment during the survey did not show significant group differences [warmth feelings: M altruistic = 4.40, SD = 1.08, M non-altruistic = 3.97, SD = 1.44, t (77) = 1.52, p = 0.134; environmental temperature: M altruistic = 69.55, SD = 5.99, M non-altruistic = 67.39, SD = 10.07, t (76) = 1.17, p = 0.246].

In summary, Study 1 primarily demonstrated the effect of altruistic behavior on warmth feelings in a retrospective real-world crisis. It is worth noting that the results for warmth feelings of the current environment did not reach significance, although in the same direction as for the remembered environment. This seemed not consistent with previous research when recalling a past experience of social exclusion affected current feelings of warmth (e.g., Zhong and Leonardelli, 2008 ). This was possibly due to individuals’ separation of the current feelings and remembered feelings. In addition, the study was less controlled due to the nature of the real crisis scenario. Thus, well-controlled experiments were conducted in the laboratory by measuring and manipulating altruistic behaviors, and warmth feelings were collected instantly after the behaviors.

Study 2a was designed to replicate the effect of altruistic behavior on warmth perception of the ambient environment in crisis context with three improvements. First, we created an earthquake scene in the laboratory. Participants were asked to imagine that they were stuck in an earthquake scene in which a stranger was requesting help. Second, the temperature of this laboratory-based earthquake scene was kept stable at 15°C. Third, altruistic willingness was directly measured in this study so that we could observe the relationship between altruistic willingness and warmth feelings. To match the settings of the scenario, participants took the experiment in groups of 4 and there was no former acquaintance within each group. Such design could increase a sense of reality and offer a closer simulation of the scenario. It could also be more natural to induce helping behaviors in such conditions. We expected that the participants with higher altruistic willingness would perceive higher levels of warmth of the ambient environment.

Because of the correlational nature of Study 2, a power of 0.8 and a medium level of r ( r = 0.3) were used to compute the expected sample size ( Cohen, 1992 ). This resulted in an expected sample size of 67. Study 2a took place in a university in Beijing, China. We recruited 69 college students (24 males; M a ge = 22.31 years, SD = 2.97) from the campus online forum. All participants read the informed consent document and agreed to participate in the experiment for a payment of 10 RMB (approximately $1.5).

As mentioned before, the procedure was implemented in groups of four participants with no former acquaintances. They were first guided into a preparation room where they read a brief introduction of the experiment and then led to the experiment room. The experiment room was set as a scene after an earthquake, with a constant temperature of 15°C and all lights turned off. Participants were asked to sit in a circle on the floor during the whole experiment. They each were approximately 40–50 cm apart from one another.

After ensuring that each of the four participants was ready, the laptops started synchronously to play a video containing standardized instructions leading participants through the whole experiment. Participants were asked not to interact with other participants. The first part of the video was a 40-s video clip of a real earthquake. Participants were instructed to imagine that they had just experienced the earthquake and were currently stuck in a room under a collapsed building with several tourists.

In the second part, the video displayed a questionnaire with several slides, and participants were asked to write down their responses on the paper. First, they were required to report how was their perception of the ambient environment along an 11-point scale (0 = extremely cold , 10 = extremely warm ). This was served as a pre-altruism measure of warmth perception. Perceived severity and uncertainty of the situation were also measured as control variables. Second, participants were told that “one of the tourists did not have food and had requested help.” Participants were asked to “indicate (with a whole number between 0 and 100) the percentage of food they would like to share with the tourist.” The willingness-to-help was regarded as participants’ subjective willingness of altruistic behavior. Additionally, warmth perception of the ambient environment was again measured as a post-altruism measure. Finally, participants were debriefed and then paid for their participation.

To control for individual differences in feelings of warmth before altruistic behaviors, we subtracted the pre-altruism measure from the post-altruism measure of warmth perception and used this result in analyses. Regression analysis showed that willingness-to-help was positively related to change in warmth perception of ambient environment, F (1,67) = 4.39, p = 0.040, β = 0.25. This relationship remained significant when perceived severity and uncertainty of the scenario were controlled in the first step of the regression, F (1,65) = 4.46, p = 0.039, β = 0.26.

Study 2a replicated an association between altruism and warmth perception of ambient environment, demonstrating that people with a greater subjective altruistic willingness would experience warmer feelings. However, limitations remained in the measurement of altruism. First, neither Study 1 (recalled altruistic behavior) nor Study 2a (altruistic willingness) measured actual altruistic behavior. Second, the results of Study 2a were correlational, so a causal relationship between altruism and warmth feeling was not confirmed. Third, it remained unclear why the effect of altruism on perceived physical warmth emerged.

Study 2b aimed at confirming the causal relationship between altruism and warmth perception in a crisis situation. Participants were randomly manipulated to exhibit an altruistic or a non-altruistic behavior in an experiment. We also explored a possible mediator of the relationship, hypothesizing that participants in the altruistic group (vs. the non-altruistic group) would report a warmer perception of the ambient environment because they felt a closer social distance to the help seeker. Moreover, we measured participants’ prosocial traits to control for the possible confounding effects of individual differences.

The computation of the sample size was identical to that of Study 1, resulting in a sample size of 72. Study 2b took place in a university in Beijing, China. Eighty-five university students were recruited from the campus online forum and randomly assigned to one of the two conditions (altruistic vs. non-altruistic). Five participants were excluded from the final data analysis because they responded to less than two-thirds of the assessment items. The final sample (28 males; M a ge = 22.41 years, SD = 3.04) comprised 36 participants in the altruistic condition and 44 participants in the no-altruistic condition. Participants each received a payment of 10 yuan RMB (approximately $1.5).

The settings and procedure of Study 2b were almost identical to those of Study 2a except for several important changes. First, although participants still participated in groups of four, one of the four participants in each group was actually a confederate.

Second, in addition to reading a brief introduction of the experiment in the preparation room, each participant was given a bag. They were told not to open and examine the bag until the instructions told them to do so in the experiment. The three bags for the real participants each contained ten small packs of bread (20 g each). There was no food in the confederate’s bag.

Third, the participants were guided to the experiment room, which had the same settings as in Study 2a. But after watching the first part of the video, the participants were instructed to open the bag and count the bread packs. Because participants were sitting close to each other (∼40–50 cm away from each other), they could easily notice that the confederate had nothing in the bag.

Fourth, participants were then instructed to report (1) the pre-altruism measure of warmth perception of the ambient environment and (2) control variables. Apart from the perceived severity and uncertainty of the situation, negative emotions (i.e., sadness, fear, desperation, and stress; α = 0.85) were also included as control variables.

Fifth, after the above measures were taken, participants were told that the person with no food in bag had requested help. Moreover, participants were manipulated into two groups by reading the following words: ‘ gathering everyone’s power and sharing resources with the group were the best choice in crisis’ (altruistic group) or ‘ saving one’s own resources and maximizing one’s own benefits were the best choice ’ (non-altruistic group). Each participant was asked to decide how many bread packs he/she would like to give to the help-seeker and to take the corresponding number of packs out of the bag. Participants were also instructed to write down the number on the answer sheet. This was used as the manipulation check of altruistic behavior.

Sixth, participants were asked to report perceived psychological distance with the other tourists in the situation on a 7-point scale (1 = close , 7 = distant ) and this measure of social distance was used as the proposed mediator. The post-altruism measure of warmth perception was measured identically to Study 2a. Finally, participants completed the Social Value Orientation Questionnaire as a measure of prosocial traits after returning to the preparation room.

Prosocial Trait

We used the Social Value Orientation Questionnaire ( Van Lange et al., 1997 ) to measure prosocial traits. This questionnaire contains nine multiple-choice situations. In each situation, a participant needed to choose one of three options to decide the outcomes for himself/herself and for another player. The three options in each situation corresponded to three social value orientations: competitive (seeking a larger difference between one’s own and other’s outcomes), individualistic (seeking a larger outcome for oneself), and prosocial (seeking a larger joint outcome). If the choices of six or more out of the nine situations were consistent with one of these social value orientations, participants were classified accordingly. Otherwise, the participant was designated as unclassified.

The manipulation check of altruistic behavior was successful. Participants in the altruistic condition shared more packs of bread ( M = 3.11, SD = 0.98) with the help-seeker than those in the non-altruistic condition ( M = 2.55, SD = 1.13), t (78) = 2.36, p = 0.021, d = 0.53.

In terms of change in warmth perception of the ambient environment, the result was consistent with Studies 1 and 2a. Participants in the altruistic group reported more increase in warmth feelings of the ambient environment ( M = 0.44, SD = 2.34) than those in the non-altruistic group ( M = -0.70, SD = 2.53), t (78) = 2.09, p = 0.040, d = 0.47. The effect remained significant when negative emotions, perceived severity and uncertainty of the scenario were controlled as covariates in the analysis, F (1,74) = 4.33, p = 0.041, η 2 = 0.06. 2

A Chi-square test showed that there was a tendency that the distributions of the four social value orientations (competitive, individualistic, prosocial, and unclassified) in the two groups were not balance [χ 2 (3) = 6.43, p = 0.092]. We also tested whether social value orientations affect change in warmth perception. Because social value orientation was a categorical variable, we run a 2 (experimental manipulation: altruistic vs. non-altruistic) × 4 (social value orientation: competitive vs. individualistic vs. prosocial vs. unclassified) ANOVA. Because there was no significant interactions between the experimental manipulation and the social value orientation ( p = 0.524), a custom model was run for analyzing only the main effects. Results showed that the effect of experimental manipulation remained marginally significant, F (1,75) = 3.39, p = 0.070, η 2 = 0.04. And social value orientation did not significantly affect change in warmth perception, F (3,75) = 1.50, p = 0.232, η 2 = 0.06.

Next, we examined whether perceived social distance mediated these effects. Participants in the altruistic group reported a significantly shorter perceived distance ( M = 2.53, SD = 1.00) than those in the non-altruistic group ( M = 3.16, SD = 1.46), t (78) = 2.20, p = 0.031, d = 0.50. Regression analysis showed that the group variable (the altruistic group coded as “1” and the non-altruistic group coded as “0”) was a significant predictor of warmth perception (β = 0.23, p = 0.040). However, the strength of this relationship became non-significant (β = 0.18, p = 0.105) when perceived social distance was included in the analysis. Moreover, perceived social distance had a tendency to be negatively related to warmth perception (β = -0.19, p = 0.091). The above results imply a mediation effect of perceived social distance. To further confirm the mediation effect, a 5000-sample bootstrapping analysis was conducted. Results showed that the 95% bias-corrected confidence interval for the indirect effect was [0.03, 0.61], suggesting a significant indirect effect ( MacKinnon et al., 2007 ). Therefore, the results of Study 2b confirmed that engaging in altruistic behaviors can increase the perceived warmth of the ambient environment while also suggesting that reduced social distance could be an internal process.

So, the first three studies focused on the relationship between altruism (recalled altruistic behavior, altruistic willingness, and manipulated altruistic behavior) and perceived warmth of the ambient environment. The scenarios used in these studies were all crisis situations with low environmental temperatures. The temperature setting was important because a cold environment could be an extreme threat to survival during a crisis. Thus, regulating warmth perception of the ambient environment would be of great significance.

In the following three studies, we aimed to extend these findings to non-crisis situations with cold environments. This was mainly for two reasons. First, although we explored altruistic behaviors with either real or laboratory-based crisis situations, the manipulation and observation of altruistic behaviors suffered from some limitations. We could obtain only retrospective data from real crisis, and the manipulation of altruistic behaviors was direct. Second, participants in Studies 2a and 2b took the experiments in groups, which could result in interdependence of data. Although we did not find significant influence of group in Study 2b, following studies are aimed to avoid this problem by changing the experimental design. Third, the situations used in the previous studies were crisis-related, and we would like to confirm the relationship in less-threatening daily situations.

Study 3a was designed to examine the association between altruistic behaviors and perceived warmth of the ambient environment in our daily life. We made two improvements in this study. First, we adopted temperature estimation as an additional index of warmth perception. Second, body temperature was measured as a control variable. We predicted that participants who chose to behave altruistically would report feeling warmer in the environment than those who chose not to help. In this study, we also predicted a mediation effect of perceived social distance.

The computation of the sample size was identical to that of Study 1, resulting in a sample size of 72. Study 3a took place in a university in Beijing, China and participants were college students recruited from the campus online forum. The final sample contained 64 participants (33 males; M a ge = 22.66 years, SD = 2.47). Thirty-two participants who engaged in the altruistic task were labeled the altruistic group and the rest were labeled the non-altruistic group.

The temperature of the experiment room was maintained constantly at 15°C. Seventy-one college students were recruited and completed a 10-min irrelevant decision-making questionnaire for a reward of 10 yuan RMB (approximately $1.5). After participants were paid, they were invited to participate in an additional activity organized by the Student Union of the Department of Psychology, but for no extra reward. Seven participants refused to participate and were excluded from the final analysis. Thus, the final sample contained 64 participants.

The additional activity involved two tasks for children from low-income migrant workers’ families. These children are usually regarded as a disadvantaged group in China. The first task was time-consuming and required considerable attention. Participants needed approximately 10 min to read and revise some educational materials for these children. The second task required participants to just complete a 1-min questionnaire about their understandings on these children. In this questionnaire, a 1-item measure of perceived social distance toward these children was presented along a 7-point scale (1 = extremely close , 7 = extremely distant ) with some filling items. Participants could decide whether to participate in both tasks or in the second task only. Based on their decisions, participants were labeled as altruistic or non-altruistic.

Lastly, participants were asked to finish another small survey entitled “A survey on the Laboratory Environment.” In the survey, participants were required to report their perception of the warmth of the experiment room on an 11-point scale (0 = extremely cold , 10 = extremely warm) . They were also asked to estimate the room temperature in degrees Celsius. As a control, we measured the body temperature of the participants using a non-contact infrared thermometer. Finally, participants were debriefed and thanked for their participation.

As expected, participants in the altruistic group felt the room was warmer ( M = 6.06, SD = 2.18) than those in the non-altruistic group ( M = 4.53, SD = 2.09), t (62) = 2.86, p = 0.006, d = 0.72. This effect remained significant when body temperature was included as a covariate, F (1,61) = 9.31, p = 0.003, η 2 = 0.13. Participants in the altruistic group also reported higher estimates of the room temperature ( M = 16.91, SD = 5.21) than those in the non-altruistic group ( M = 12.58, SD = 4.48), t (62) = 3.57, p = 0.001, d = 0.89. This effect also remained significant when body temperature was included as a covariate, F (1,61) = 12.34, p = 0.001, η 2 = 0.17. Moreover, there was a significantly positive correlation between feeling of warmth and estimate of the room temperature ( r = 0.36, p = 0.002). However, body temperature did not correlate with feeling of warmth ( r = -0.20, p = 0.120) or estimate of the room temperature ( r = 0.05, p = 0.671). We then examined whether perceived social distance mediated the effects of group conditions on warmth perception. Participants in the altruistic group reported a significantly shorter distance ( M = 3.66, SD = 1.54) than those in the non-altruistic group ( M = 4.47, SD = 1.32), t (62) = 2.27, p = 0.027, d = 0.57. Regression analysis showed that the group variable (the altruistic group coded as “1” and the non-altruistic group coded as “0”) was a significant predictor of warmth perception, β = 0.34, p = 0.006). This relationship weakened (β = 0.28, p = 0.024) when perceived social distance was included in the analysis. Moreover, perceived social distance had a tendency to be negatively related to warmth perception (β = -0.21, p = 0.087). The above results implied a mediation effect of perceived social distance. To further confirm the mediation effect, a 5000-sample bootstrapping analysis was conducted. Results showed that the 95% bias-corrected confidence interval for the indirect effect was [0.04, 0.89], suggesting a significant indirect effect ( MacKinnon et al., 2007 ).

In conclusion, the effect of altruistic behavior on perceived warmth and the mediating effect of perceived social distance were verified in a daily life situation. In Study 3b, we imported a stricter control of altruistic behaviors to confirm the effect of altruistic behavior on warmth perception of the ambient environment.

In Study 3b, we randomly assigned participants to an altruistic or a non-altruistic condition. Moreover, a no-task group was added as a control group to rule out the alternative explanation that non-altruistic behavior could lead to a decrease in warmth perception of the ambient environment. We hypothesized that participants in the altruistic group would report higher warmth perception than those in the non-altruistic group and the no-task group. We expected participants in the non-altruistic and the no-task groups to report equal feelings of warmth.

Because Study 3b was designed to be consisted of three groups, a power of 0.8 and an effect size f of 0.35 were used to compute the expected sample size, resulting in a sample size of 81. Study 3b took place in a university in Beijing, China. Eighty-three college students were recruited from the campus online forum. Eight of them did not finish the experiment and were thus excluded from the analysis. The final sample contained 75 participants (36 males; M a ge = 22.2 years, SD = 7.42). Participants were randomly assigned to one of the three conditions, resulting in 25 participants in each group.

Participants were led to the experiment room, the temperature of which was maintained constantly at 15°C. In the altruistic condition, participants were invited to take part in a charity activity called ‘The Love Crossing 4000 Kilometers’ to help the students from a remote town called Jimunai. This town was one of the poorest areas in China, located 4000 km from Beijing. The charity aimed to collect stationery for these students and offer them more opportunities to learn about the outside world. Each participant was invited to write a postcard for a particular student in Jimunai Secondary School to introduce an attraction in Beijing. Participants then were given a 10-RMB (approximately $1.5) cash note and were required to put the note in a charity box. Notably, all of the postcards were sent to the corresponding students, and the money was donated to the school in the name of participants after the experiment.

In the non-altruistic condition, participants were asked to read information about Beijing’s candidature for the 2022 Winter Olympic Games. Participants were told that the laboratory was helping to advocate the attractions in Beijing. They could choose one postcard and write down an introduction about an attraction in Beijing. In the no-task group, participants were assigned no additional task.

All the participants then were asked to complete the survey about the laboratory environment, which was identical to that used in Study 3a. In the survey, participants were required to report their warmth perception of the experiment room along an 11-point scale (0 = extremely cold , 10 = extremely warm ). As a control, we measured the body temperature of the participants using a non-contact infrared thermometer. Finally, all the participants were debriefed, thanked, and each paid 10 RMB (approximately $1.5) for their participation.

To ensure the successful manipulation of altruism, we asked another 78 participants (36 males) to read the scenarios in Study 3b and to then indicate the extent of altruism (‘To what extent do you feel you could help others if you participate in this activity?’ 1 = not at all , 5 = great deal ). Results showed that the participants in the altruistic condition reported more helping ( M = 3.63, SD = 0.75) than those in the non-altruistic condition ( M = 3.18, SD = 0.90), t (76) = 2.42, p = 0.018, d = 0.55. To exclude the possibility that the non-altruistic material would induce thoughts of competition, the participants also indicated perceived competition (‘To what extent do you feel you are competing with others?’ 1 = not at all , 5 = great deal ). No significant difference was found between the participants in the two conditions, p = 0.976. In addition, participants were required to write down 5–10 words that the reading material reminded them. We asked two experimenters who did not know the intention of the study to classify the words. Results showed that participants who read the altruistic scenario wrote down more altruistic-related words ( M = 1.06, SD = 0.17) than those who read the non-altruistic scenario ( M = 0.27, SD = 0.04), t (76) = 9.63, p < 0.001, d = 6.56. Similar results were found for caring-related words ( M altruistic = 1.13, SD = 0.70, M non-altruistic = 0.08, SD = 0.27, t (76) = 8.85, p < 0.001, d = 2.02.

One-way ANOVA revealed that the warmth perception of the participants were significantly different across three conditions, F (2,72) = 4.77, p = 0.011, η 2 = 0.12. As expected, Bonferroni post hoc analysis showed that participants in the altruistic group ( M = 6.84, SD = 1.89) felt significantly warmer in the room than those in the non-altruistic group ( M = 5.32, SD = 2.01) (95% CI of mean difference [0.19, 2.85], p = 0.019). Likewise, participants in the altruistic group felt significantly warmer than those in the no-task group ( M = 5.48, SD = 1.83) (95% CI of mean difference [0.03, 2.69], p = 0.042). However, there was no significant difference in warmth perception between the non-altruistic and no-task groups (95% CI of mean difference [-1.17, 1.49], p = 1.00). Moreover, the effect of condition remained significant when the body temperature of participants was included as a covariate, F (2,71) = 4.55, p = 0.014, η 2 = 0.11. And there was no significant correlation between feelings of warmth and body temperature of the participants, r = 0.08, p = 0.490.

With stronger causal inference in this instance, the results replicated the effect of altruism on the warmth perception of the ambient environment. By adding a no-task control group, this study showed that the effects on warmth perception was caused by an increase in warm feelings from behaving altruistically rather than a decrease from not behaving altruistically.

Studies 3a and 3b replicated in the laboratory the primary findings of Studies 2a and 2b with regard to daily life situations. Compared with crisis situations, daily situations were more naturally created. In Study 3a, altruistic behaviors were chosen voluntarily by participants whereas in Study 3b, participants were randomly led to believe that they did an altruistic behavior or a non-altruistic behavior. The results provided convergent evidence for the proposed effect that altruistic behaviors would increase the perception of ambient warmth. To note, Studies 3a and 3b were underpowered by 8 and 6 participants compared to the computed sample sizes, respectively. This was due to experimental constraints of unexpected loss of participants and was a limitation for both studies. To further confirm the relationship and to foster external validity, we conducted another study in a real life situation to explore how a common altruistic behavior, donation, affected warmth perception of the ambient environment.

Study 4 was a field experiment conducted to increase the external validity of previous work and confirm the effects of altruistic behavior on the perception of warmth of the ambient environment. The experiment was conducted with the help of the Students Association of Sunshine Volunteers at Peking University. An actual charity event was held on December 7, 2012, to collect donations for children with leukemia. A donation desk was placed on the sidewalk along the campus main street. We randomly solicited individuals to complete a short questionnaire after they passed the donation desk. We hypothesized that people who donated (the altruistic group) would feel warmer regarding the ambient environment than those who did not donate (the non-altruistic group).

Because Study 4 was a field study, a power of 0.8 and an effect size d of 0.5 (which was smaller than previous studies) were used to compute the expected sample size, resulting in a sample size of 102. A total of 108 people participated in this experiment (47 males; M a ge = 22.27 years, SD = 5.50). Of these, 55 made a donation.

Data were collected from 9:00 a.m. to 1:00 p.m. during the day. The environment temperature during the experiment ranged from -7°C to 0°C. Participants filled in a short questionnaire called “Perception of Weather Conditions in Beijing,” which included items assessing warm feelings about the environment along an 11-point scale (0 = extremely cold , 10 = extremely warm) and an estimation of the environment temperature in Celsius degrees. To control for hourly temperature variations, we paired a passer-by who did not donate within 2 min after a donor completed the questionnaire.

Consistent with our prediction, results showed that the passers-by who made a donation (altruists) perceived the ambient environment as warmer ( M = 3.07, SD = 2.01) than those who did not ( M = 2.40, SD = 1.86). The difference was marginally significant, t (106) = 1.81, p = 0.073, d = 0.35. The donors also reported significantly higher temperature estimations ( M = -0.51°C, SD = 3.40) compared to the non-donors ( M = -2.34°C, SD = 3.33), t (106) = 2.83, p = .006, d = 0.54).

A Meta-Analysis

So far, we have consistently revealed the effects of altruistic behavior on physical warmth. For further verification, we conducted a meta-analysis to test the statistical replication of the experiments. The present experiments provided a good condition for such an analysis for two reasons. First, the altruistic behaviors varied in category, including sharing food with others, charitable helping, and monetary donations. Second, both college students in China and residents in America were included in the experiments, offering a significant diversity of participant populations.

Because Study 2a did not involve grouping participants, data from all other five studies were included. We used warmth perception of the ambient environment as the main dependent variable. Using Comprehensive Meta Analysis software, we entered the means, standard deviations, and sample sizes of the altruistic group and the non-altruistic group to calculate effect sizes. As shown in Table ​ Table1 1 , the combined z -value was 5.94 ( p < 0.001), which confirmed that altruistic behaviors increased physical warmth. Moreover, the heterogeneity test showed that the q -value was 6.24 ( p = 0.182), indicating that the effects did not differ significantly across the studies.

Results of a mini meta-analysis of the six studies.

General Discussion

Through six experiments, we found that altruistic performers were likely to feel warmer about the ambient environment than those who either refused to help or did not have a chance to help. Furthermore, results revealed that this effect was mediated by perceived social distance (Studies 2b and 3a). Moreover, this effect was confirmed in both crisis and ordinary situations amidst different experimental settings. Altruism is a complicated construct. It is usually directed by voluntary motivations and thus altruistic behaviors were just observed in Studies 2a, 3a, and 4. To confirm the causal relationship, recalled altruistic experiences as well as instant altruistic behaviors were manipulated in Studies 1, 2b, and 3b. Moreover, different instructions and cover stories were incorporated in Studies 2b and 3b. In summary, this design of series studies complemented for possible weaknesses in terms of constructing altruism and offered convergent findings.

Immediate Self-Reward of Altruistic Behaviors

While improving the welfare of others, altruistic behaviors typically deplete the energy and resources of the altruistic performers. Earlier explanations of altruism focused on its long-term return, either through genetic propagation ( Hamilton, 1963 ; Hamilton and Axelrod, 1981 ) or reciprocity ( Trivers, 1971 ; Nowak and Sigmund, 1998 , 2000 ). However, research on altruistic punishment observed the activation of the reward system of the dorsal striatum, signaling an instant satisfaction from altruistic behaviors ( de Quervain et al., 2004 ). Results from the current research have implied a potential self-reward mechanism for altruistic behaviors.

We found an immediate rewarding effect on the perceived warmth of the ambient environment after performing altruistic behaviors. In six studies, warm feelings were measured immediately after the participants reported the willingness to help and after they exhibited actual altruistic behaviors. This suggested that the increased feeling of warmth surrounding the ambient environment was an immediate reward rather than a long-term return-benefit for altruistic behavior performers. Furthermore, this effect was found to be mediated by perceived social distance, introducing a potential psychological process in this immediate reward. As mentioned in the introduction, previous research also found that altruistic behavior could result in some immediate psychological reward, including happiness ( Dunn et al., 2008 ), self-efficacy ( Midlarsky and Kahana, 1994 ), and positive self-evaluation ( Post, 2005 ). Moreover, Dawans et al. (2012) found that exposure to acute social stress could increase prosocial responses, which implied an immediate protective function of altruism in coping with acute stress. In the current research, the immediate reward was found to extend to concrete physical feelings through a psychological process.

We proposed three characteristics of the immediate reward system of altruism that were different from those of its long-term benefits. First, an immediate reward is a much more certain and spontaneous effect compared with long-term benefits. An altruistic performer may end up with no real benefits in the long run but could always obtain positive feedbacks through the immediate reward system. Second, the immediate reward could operate on the psychological or even perceptual level. Compared with the external benefits from genetic propagation and reciprocity, the internal immediate reward could serve as a direct incentive to engage in altruistic behaviors. Third, receiving an immediate reward could be a self-feedback process. For long-term benefits, altruistic performers usually need to rely on the behaviors or the survival possibilities of others. In contrast, altruistic behaviors could directly activate the performer’s psychological and physical processes on the individual level. To summarize, an immediate reward from altruism could be certain, internally perceived, and self-activated, compared to long-term benefits.

The immediate reward for altruistic performers could be very valuable. The observed results of altruistic behaviors in previous research included positive emotions and positive self-cognitions, all of which were found to be positively related with people’s subjective well-being (e.g., Bandura, 1986 ; Brunstein, 1993 ). A meta-analysis found that volunteers had lower risks of mortality ( Jenkinson et al., 2013 ). In the current research, we observed a consistent and direct connection between physical warmth and altruism. Physical warmth may be an important source of psychological energy or at least a strong comfort for individuals who were exposed to a cold environment. Specifically, individuals could perceive the environment as more secure and predictable ( IJzerman et al., 2015a ), and thus gain more confidence in coping with the environment. Thus, if the immediate reward from altruism could be valuable for the altruistic performers, they can be trusted because the reward would be more direct and certain.

In all, it suggests that altruism can not only bring long-term benefits to the performers (as explained in kin selection and reciprocity accounts) but also evoke immediate positive feelings inside the performers (as revealed in the current research). These two systems offer a better understanding of the functional adaptabilities of altruistic behaviors. Traditional economic analysis of altruism is based on a cost/utility analysis of external factors such as money, time, and probabilities of genetic propagations. We proposed to add internal utility (psychological states such as warmth perception and emotion) in the discussion of altruism. Altruistic behaviors would be evolutionarily meaningful if the increase in the internal utility offsets the reductions in external utility. Thus, the framework of altruism would be more complete when we concern both long-term and immediate benefits, as well as both external and internal utilities.

Lingering Fragrance Effect

A potential immediate self-reward system of altruistic behaviors has practical meaning in promoting people’s well-being and quality of life in both daily life and times of crisis. The effect found in the current research was a typical example of the interaction between an individual’s physical and psychological system. The saying the roses in her hand, the flavor in mine could reveal a real psychological activation effect, through which people could promote their personal physical states. Thus, we named this psychological activation effect as a lingering fragrance effect.

When facing a threat (e.g., hunger, pain, or cold), there are mainly two strategies for its removal. One is to cope with the threat immediately (e.g., put on a coat to remove coldness). The other strategy is to activate a psychological process to change the perception of the threat. As in the current research, altruistic behavior activated people’s psychological processes and changed their warmth feelings of a cold environment. During crisis, the adverse external conditions could hardly be changed because of insufficient support of food, water, and living conditions. However, individual cognitions and behaviors are pliant and controllable, offering possibilities to attenuate the threats in crisis. Hence, psychological activations or subjective adjustments could become a prospective coping strategy under such circumstances. Importantly, the experimental settings used to induce people’s real altruistic behaviors might be artificial in the current research. Moreover, we found consistent lingering fragrance effects in spite of this shortcoming. Thus we expect in real life that the lingering fragrance effect of altruistic behaviors should be amplified to benefit individuals in a more extensive manner.

To note, we focused on the effects of altruistic behaviors on individuals’ subjective feelings of warmth in the current study. And we did not found significant similar effects on individuals’ objective physical states (body temperature in Studies 3a and 3b). This also implied that the psychological reactions may be faster and more flexible than physical reactions. However, it is possible that the physical states could get feedbacks from psychological states when the timeline is extended concerning the close interactions of physical and psychological states (e.g., Edwards and Cooper, 1988 ) and the prospective treatment of biofeedback (e.g., Lagos et al., 2013 ). In addition, we have mentioned in the introduction as well as in the discussion section that an increase in warmth feelings could be helpful for individuals who were coping with cold environment. Warmth feelings could serve as an easily accessible comfort for them especially in some crisis situations. On the other hand, it could become a problem when individuals optimistically perceived the environment as warm but failed to respond to the important environmental cue about temperature. This indicates that a raise of warmth feelings may actually become a double-edge sword in some cases. And it is interesting and prospective to explore the other edge of the sword in future studies.

In summary, the lingering fragrance effect is an important perspective for understanding the strategy in coping with a threat. The interaction between physical and psychological systems makes it possible to change the physical state through the mobilization of internal psychological resources. Under the crisis circumstances with limited conditions, using a psychological resource to resist the bad impacts of a crisis could be a more reasonable or even a single possible strategy to cope with threats. As the results showed, active altruistic behavior is a significant way for people to resist the cold environment. According to the Harry Truman effect, a person’s potential can be activated under certain circumstance ( Seligman, 2002 ). Equally, a person can activate his/her own internal psychological potential to cope with the intense threat and pursue positive results.

Future Directions

The exploration of the proposed immediate self-reward model of altruistic behaviors could be expanded from two aspects. First, the crisis situations and the experimental settings in the current research were all related to coldness. In some cases, coldness could even be a significant threat to survival (e.g., Hurricane Sandy in Study 1 and the earthquake in Studies 2a and 2b). In such cases, increased feelings of ambient warmth were regarded as a reward for the individuals. The specific reward and the corresponding psychological processes might change with different adverse situations. For example, coolness could become a more comfortable state when the environment was extremely hot. Future research could help to explore different contents of the lingering fragrance effect and to offer more evidence for an immediate internal reward system.

Second, further studies could be conducted on the rewarding nature of altruism. For example, a positive cycle might be established in which increased warmth perception following altruistic behaviors result in future altruistic behaviors. Moreover, de Quervain et al. (2004) found that effective altruistic punishment was related to stronger activation of the dorsal striatum. Similarly, neuroscientific studies would be helpful in finding out whether performing an altruistic behavior would activate the same region or some other reward-related brain regions.

Ethics Statement

The research project was approved by the Ethics Committee of Department of Psychology, Peking University. Studies 3a, 3b, and 4 involved helping disadvantaged groups (i.e., children from low-income migrant workers’ families, students from a remote and poor town, and children with leukemia). All of the helping scenarios used in the studies (including refining education materials, sending postcards, and donations) were actually carried out.

Author Contributions

T-YH, JL, HJ, and XX should be considered as joint first authors as they contributed equally to the research.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

We would like to appreciate Xiaoxiao Hu for helping us with data collection. Also, many thanks to Student Union of Psychology in Peking University, Jimunai Middle School and Qizhi Ning, who offered us opportunities to conduct Studies 3a, 3b, and 4 and to help those kids who were in need. Last but not least, we would like to extend our sincere gratitude to the members of Risk and Decision-making Lab of Psychology Department of Peking University for their helpful comments and support on the research.

Funding. This work was supported by the Key Program of National Natural Science Foundation of China [grant number 91224002]; and the General Program of National Nature Science Foundation of China [grant number 71172024].

1 http://www.mturk.com

2 Participants in Study 2b took the experiment in group of 4 (three participants and one confederate) and they needed to act to share food during the experiment. Although participants were asked not to interact with others, it was possible that their reactions were influenced by other participants. To rule out this effect, we run an HLM test with a contextual model. The average number of packs within each group was regarded as the contextual variable. Results showed that the effects of altruism manipulation was marginally significant, t (77) = 1.84, p = 0.069. The effect of contextual variable did not reached significance, t (77) = 0.18, p = 0.860. The results remained when negative emotions, perceived severity and uncertainty of the scenario were included in the model.

• Bandura A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall. [ Google Scholar ]
• Bargh J. A., Shalev I. (2012). The substitutability of physical and social warmth in daily life. Emotion 12 154–162. 10.1037/a0023527 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Barsalou L. W. (1999). Perceptual symbol system. Behav. Brain Sci. 22 577–609. 10.1017/S0140525X99532147 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Barsalou L. W. (2008). Cognitive and neural contributions to understanding the conceptual system. Curr. Dir. Psychol. Sci. 17 91–95. 10.1111/j.1467-8721.2008.00555.x [ CrossRef ] [ Google Scholar ]
• Basbaum A. I., Fields H. L. (1984). Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Annu. Rev. Neurosci. 7 309–338. 10.1146/annurev.ne.07.030184.001521 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Batson C. D. (1991). The Altruism Question: Toward a Social-Psychological Answer. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. [ Google Scholar ]
• Beckes L., Coan J. A. (2011). Social baseline theory: the role of social proximity in emotion and economy of action. Soc. Pers. Psychol. Compass 5 976–988. 10.1111/j.1751-9004.2011.00400.x [ CrossRef ] [ Google Scholar ]
• Boyd R., Gintis H., Bowles S., Richerson P. J. (2003). The evolution of altruistic punishment. Proc. Natl. Acad. Sci. U.S.A. 100 3531–3535. 10.1073/pnas.0630443100 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Brunstein J. C. (1993). Personal goals and subjective well-being: a longitudinal study. J. Pers. Soc. Psychol. 65 1061–1070. 10.1037/0022-3514.65.5.1061 [ CrossRef ] [ Google Scholar ]
• Chamarro A., Fernández-Castro J. (2009). The perception of causes of accidents in mountain sports: a study based on the experiences of victims. Accid. Anal. Prev. 41 197–201. 10.1016/j.aap.2008.10.012 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Cohen J. (1992). A power primer. Psychol. Bull. 112 155–159. 10.1037/0033-2909.112.1.155 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Danielli J. F. (1980). Altruism and the internal reward system or the opium of the people. J. Soc. Biol. Struct. 3 87–94. 10.1016/0140-1750(80)90001-9 [ CrossRef ] [ Google Scholar ]
• Dawans B., Fischbacher U., Kirschbaum C., Fehr E., Heinrichs M. (2012). The social dimension of stress reactivity: acute stress increases prosocial behavior in humans. Psychol. Sci. 23 651–660. 10.1177/0956797611431576 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• de Quervain D. J. F., Fischbacher U., Treyer V., Schellhammer M., Schnyder U., Buck A., et al. (2004). The neural basis of altruistic punishment. Science 305 1254–1258. 10.1126/science.1100735 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• de Waal F. B. M. (2008). Putting the altruism back into altruism: the evolution of empathy. Annu. Rev. Psychol. 59 279–300. 10.1146/annurev.psych.59.103006.093625 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Delgado M. R., Locke H. M., Stenger V. A., Fiez J. A. (2003). Dorsal striatum responses to reward and punishment: effects of valence and magnitude manipulations. Cogn. Affect. Behav. Neurosci. 3 27–38. 10.3758/CABN.3.1.27 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Dunn E. W., Aknin L. B., Norton M. I. (2008). Spending money on others promotes happiness. Science 319 1687–1688. 10.1126/science.1150952 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Edwards J. R., Cooper C. L. (1988). The impact of positive psychological states on physical health: a review and theoretical framework. Soc. Sci. Med. 27 1447–1459. 10.1016/0277-9536(88)90212-2 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gintis H. (2000). Strong reciprocity and human sociality. J. Theor. Biol. 206 169–179. 10.1006/jtbi.2000.2111 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Gintis H., Bowles S., Boyd R., Fehr E. (2003). Explaining altruistic behavior in humans. Evol. Hum. Behav. 24 153–172. 10.1016/S1090-5138(02)00157-5 [ CrossRef ] [ Google Scholar ]
• Hamilton W. D. (1963). The evolution of altruistic behavior. Am. Nat. 97 354–356. 10.2307/2458473 [ CrossRef ] [ Google Scholar ]
• Hamilton W. D., Axelrod R. (1981). The evolution of cooperation. Science 211 1390–1396. 10.1126/science.7466396 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Harbaugh W. T. (1998). What do donations buy? A model of philanthropy based on prestige and warm glow. J. Pubic Econ. 67 269–284. 10.1016/S0047-2727(97)00062-5 [ CrossRef ] [ Google Scholar ]
• Harlow H. F. (1958). The nature of love. Am. Psychol. 13 673–685. 10.1037/h0047884 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Heir H., Piatigorsky A., Weisæth L. (2009). Longitudinal changes in recalled perceived life threat after a natural disaster. Br. J. Psychiatry 194 510–514. 10.1192/bjp.bp.108.056580 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Henrich J., McElreath R., Barr A., Ensminger J., Barrett C., Bolyanatz A., et al. (2006). Costly punishment across human societies. Science 312 1767–1770. 10.1126/science.1127333 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• IJzerman H., Coan J. A., Wagemans F. M., Missler M. A., van Beest I., Lindenberg S., et al. (2015a). A theory of social thermoregulation in human primates. Front. Psychol. 6 : 464 10.3389/fpsyg.2015.00464 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ijzerman H., Gallucci M., Pouw W. T., Weibetagerber S. C., Van Doesum N. J., Williams K. D. (2012). Cold-blooded loneliness: social exclusion leads to lower skin temperatures. Acta Psychol. 140 283–288. 10.1016/j.actpsy.2012.05.002 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• IJzerman H., Janssen J. A., Coan J. A. (2015b). Maintaining warm, trusting relationships with brands: increased temperature perceptions after thinking of communal brands. PLoS ONE 10 : e0125194 10.1371/journal.pone.0125194 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• IJzerman H., Semin G. R. (2009). The thermometer of social relations: mapping social proximity on temperature. Psychol. Sci. 20 1214–1220. 10.1111/j.1467-9280.2009.02434.x [ PubMed ] [ CrossRef ] [ Google Scholar ]
• IJzerman H., Semin G. R. (2010). Temperature perceptions as a ground for social proximity. J. Exp. Soc. Psychol. 46 867–873. 10.1016/j.jesp.2010.07.015 [ CrossRef ] [ Google Scholar ]
• Inagaki T. K., Eisenberger N. I. (2013). Shared neural mechanisms underlying social warmth and physical warmth. Psychol. Sci. 24 2272–2280. 10.1177/0956797613492773 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Jenkinson C. E., Dickens A. P., Jones K., Thompson-Coon J., Taylor R. S., Rogers M., et al. (2013). Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers. BMC Public Health 13 : 773 10.1186/1471-2458-13-773 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kang Y., Williams L. E., Clark M. S., Gray J. R., Bargh J. R. (2011). Physical temperature effects on trust behavior: the role of insula. Soc. Cogn. Affect. Neurosci. 6 507–515. 10.1093/scan/nsq077 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kasperson R. E., Renn O., Slovic P., Brown H. S., Emel J., Goble R., et al. (1988). The social amplification of risk: a conceptual framework. Risk Anal. 8 177–187. 10.1111/j.1539-6924.1988.tb01168.x [ CrossRef ] [ Google Scholar ]
• Kellogg B. (2013). Sandy’s Snowy Side Turns Deadly. Available at: http://www.weather.com/storms/hurricane/news/sandy-snowy-side-20121030 [ Google Scholar ]
• Knutson B., Westdorp A., Kaiser E., Hommer D. (2000). FMRI visualization of brain activity during a monetary incentive delay task. Neuroimage 12 20–27. 10.1006/nimg.2000.0593 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Lagos L., Thompson J., Vaschillo E. (2013). A preliminary study: heart rate variability biofeedback for treatment of postconcussion syndrome. Biofeedback 41 136–143. 10.5298/1081-5937-41.3.02 [ CrossRef ] [ Google Scholar ]
• Landau M. K., Meier B. P., Keefer L. A. (2010). A metaphor-enriched social cognition. Psychol. Bull. 136 1045–1067. 10.1037/a0020970 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Leknes S., Tracey I. (2008). A common neurobiology for pain and pleasure. Nat. Rev. Neurosci. 9 314–320. 10.1038/nrn2333 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• MacKinnon D. P., Fairchild A. J., Fritz M. S. (2007). Mediation analysis. Annu. Rev. Psychol. 58 593–614. 10.1146/annurev.psych.58.110405.085542 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Midlarsky E., Kahana E. (1994). Altruism in Later Life. Thousand Oaks, CA: Sage. [ Google Scholar ]
• Molin J., Mellerup E., Bolwig T., Scheike T., Dam H. (1996). The influence of climate on development of winter depression. J. Affect. Disord. 37 151–155. 10.1016/0165-0327(95)00090-9 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Myers D. (1993). Social Psychology , 4th Edn New York, NY: McGraw-Hill. [ Google Scholar ]
• Nowak M. A., Sigmund K. (1998). Evolution of indirect reciprocity by image scoring. Nature 393 573–577. 10.1038/31225 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Nowak M. A., Sigmund K. (2000). Shrewd investments. Science 288 819–820. 10.1126/science.288.5467.819 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Post S. G. (2005). Altruism, happiness, and health: it’s good to be good. Int. J. Behav. Med. 12 66–77. 10.1207/s15327558ijbm1202_4 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Schubert T. W. (2005). Your highness: vertical positions as perceptual symbols of power. J. Pers. Soc. Psychol. 89 1–21. 10.1037/0022-3514.89.1.1 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Seligman M. E. (2002). Authentic Happiness: Using the New Positive Psychology to Realize Your Potential for Lasting Fulfillment. New York, NY: Simon and Schuster. [ Google Scholar ]
• Szymkow A., Chandler J., IJzerman H., Parzuchowski M., Wojciszke B. (2013). Warmer hearts, warmer rooms: how positive communal traits increase estimates of ambient temperature. Soc. Psychol. 44 167–176. 10.1027/1864-9335/a000147 [ CrossRef ] [ Google Scholar ]
• Trivers R. L. (1971). The evolution of reciprocal altruism. Q. Rev. Biol. 46 35–57. 10.2307/2822435 [ CrossRef ] [ Google Scholar ]
• Van Lange P. A. M., Otten W., De Bruin E. M. N., Joireman J. A. (1997). Development of prosocial, individualistic, and competitive orientations: theory and preliminary evidence. J. Pers. Soc. Psychol. 73 733–746. 10.1037/0022-3514.73.4.733 [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ward V. (2015). Winter Death Toll ‘to Exceed 40000’. The Telegraph. Available at: http://www.telegraph.co.uk/news/weather/11382808/Winter-death-toll-to-exceed-40000.html [ Google Scholar ]
• Webb R. (2014). Death Toll Rises from Harsh Winter Weather. USA Today. Available at: http://www.usatoday.com/story/weather/2014/01/07/cold-storms-deaths/4355329 [ Google Scholar ]
• Williams L. E., Bargh J. A. (2008). Experiencing physical warmth promotes interpersonal warmth. Science 322 606–607. 10.1126/science.1162548 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Zhong C. B., Leonardelli G. J. (2008). Cold and lonely: does social exclusion literally feel cold? Psychol. Sci. 19 838–842. 10.1111/j.1467-9280.2008.02165.x [ PubMed ] [ CrossRef ] [ Google Scholar ]