04 Apr Discuss and synthesize the materials from McCullough and colleagues. How is the research on prosocial behavior related to the literature on aggression? McCullough, M. (2008). Bey
Discuss and synthesize the materials from McCullough and colleagues. How is the research on prosocial behavior related to the literature on aggression?
- McCullough, M. (2008). Beyond revenge: The evolution of the forgiveness instinct. Jossey-Bass. Chapters 5 & 7 McCullough, M. (2008). Beyond revenge: The evolution of the forgiveness instinct. Jossey-Bass. Chapters 5 & 7 – Alternative Formats
- McCullough, M. E., & Tabak, B. A. (2010). Prosocial behavior. Advanced Social Psychology, Oxford, New York. McCullough, M. E., & Tabak, B. A. (2010). Prosocial behavior. Advanced Social Psychology, Oxford, New York. – Alternative Formats
2
1 Part 3 Core Topics in Social Psychology
08-Baumeister_Ch-08.indd 26108-Baumeister_Ch-08.indd 261 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
08-Baumeister_Ch-08.indd 26208-Baumeister_Ch-08.indd 262 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 Chapter 8
Prosocial Behavior Michael E. McCullough and Benjamin A. Tabak
Chimpanzees are our closest living relatives, with % of our genetic code over- lapping theirs (Varki & Nelson, ). Th is deep genetic similarity produces profound physical similarities—and also behavioral ones. Th ese behavioral similarities are nowhere better illustrated than in the realm of prosocial behav- ior. Chimpanzees, like humans from hunter–gatherer societies, hunt coopera- tively. Like humans, they patrol their territories in groups and engage in coordinated group violence against other groups (Wrangham & Peterson, ). Within groups, they form coalitions to defeat individuals too powerful for any of the coalition members to defeat on their own (de Waal, ), and males join forces to prevent each others’ mates from straying (Silk et al., ). Individuals also make an eff ort to reconcile with valuable relationship partners with whom they have recently experienced confl ict (Koski, Koops, & Sterck, ) and to comfort valuable relationship partners who have recently been the recipients of other individuals’ aggressive behavior (Fraser, Stahl, & Aureli, ). In addition, chimpanzees recognize when they need a partner to obtain a desirable food item, and they know which potential partners are likely to be most helpful to them (Melis, Hare, & Tomasello, ). Evidence also suggests that chimpanzees, like humans, will help others gain access to desired items even when they cannot immediately benefi t from a return favor (Warneken, Hare, Melis, Hanus, & Tomasello, ; Warneken & Tomasello, ).
But we cannot overlook that % uniqueness, which indicates that there are approximately million genetic diff erences between humans and chimpanzees
08-Baumeister_Ch-08.indd 26308-Baumeister_Ch-08.indd 263 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
core topics in social psychology
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 due to base pair diff erences and nucleotide additions or deletions (Varki & Nelson, ). Th at uniqueness leads to important diff erences in human and chimpanzee prosocial behavior. For example, chimpanzees show no preference for behaviors that enable others to acquire food when they are attempting to acquire food (Jensen, Hare, Call, & Tomasello, ; Silk et al., ), but such behavior is common among humans to the point of banality: If I’m going out to get lunch, I just might off er to pick something up for you. Moreover, human infants are better than chimpanzees at inferring humans’ needs and then ren- dering appropriate forms of help (Warneken & Tomasello, ). Likewise, even though both humans and chimpanzees help others in some instances, it is humans and not chimpanzees that raise armies for the common defense, seek out training so that they can render more eff ective emergency aid to others, and endure taxation to provide help for the poor and needy. Th ese important behav- ioral diff erences may refl ect fundamental qualitative diff erences in evolved cog- nitive capacities such as delay of gratifi cation (Stevens, Cushman, & Hauser, ), the ability to infer other people’s mental states from their behavior and to act empathically on the basis of that knowledge (Liszkowski, Carpenter, & Tomasello, ), and the ability to generate and learn from culture (Richerson & Boyd, ). In this chapter, we will explore some of the more interesting features of humans’ tendencies to engage in helping, sharing, and cooperating— that is, the behaviors collectively known as “prosocial behaviors” (Penner, Dovidio, Piliavin, & Schroeder, ). We will describe the classic social– psychological work on this topic, and also some of the more important recent theoretical and empirical advances, beginning with the evolutionary models that are sometimes invoked to explain humans’ prosocial tendencies.
Evolutionary Models of Prosocial Behavior
Evolutionary researchers study the body’s (brain/mind included) present struc- tures by searching for the functions those structures evolved to serve in the past: Th eir project is usually (although not always; Andrews, Gangestad, & Matthews, ) an adaptationist one (Tooby & Cosmides, ; see Maner & Kenrick, Chapter , this volume). Adaptationism relies on the fact that indi- vidual organisms within a population that vary on a trait due to genotypic diversity can incur diff ering rates of genetic propagation (i.e., fi tness) if some variants of the trait (and, therefore, the genes that contribute to their assembly during development) cause higher rates of reproduction than do others because of their ability to cause organisms to respond to specifi c adaptive challenges eff ectively. Because of these phenotype-dependent diff erences in fi tness, small
08-Baumeister_Ch-08.indd 26408-Baumeister_Ch-08.indd 264 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
Prosocial Behavior
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 incremental changes in the genes that collectively give rise to the body’s mecha- nisms (e.g., the heart, the fi ngernails, the brain’s reward circuitry) that enhance the fi tness of the bearer of those genes can gradually shape the species-typical structure of those mechanisms.
Because of natural selection’s relentless favoritism for genes that enhance their bearers’ fi tness, prosocial behavior has been an evolutionary puzzle since Darwin (/): Incurring costs (even small costs in the currencies of money, time, or energy should redound to fi tness) that benefi t someone else’s fi t- ness (e.g., when someone saves a drowning child or donates blood for a stranger’s benefi t) at fi rst glance appears to be bad evolutionary bookkeeping. Nevertheless, several evolutionary processes have been identifi ed that can help explain the evo- lution of mental mechanisms for prosocial behavior in humans (McAndrew, ; Nowak, ; Wilson & Wilson, ). Here, we focus on the models that have (we think) the greatest potential to inform social psychology: kin altruism, direct reciprocity, indirect reciprocity, signaling, and group (or multilevel) selec- tion. What makes these theories useful to social psychology is that they imply that the mind possesses specifi c functional systems that natural selection designed for their effi cacy in producing certain types of prosocial behavior. If we under- stand the selection pressureswe can formulate hypotheses about the operation of the psychological systems that evolved in response to those pressures and the social factors that activate and condition the operation of those systems.
Kin Altruism
Humans regularly endure tremendous energetic costs (e.g., gestation, nursing, feeding, sheltering, clothing, paying for college) to help their off spring and other genetic relatives. Th e theory of kin altruism explains such behaviors by exploit- ing the fact that one’s fi tness is not a function of the number of one’s off spring that survive to reproductive maturity, but rather a function of the number of off spring one has plus the number of off spring that one’sour genetic relatives have (Hamilton, ). Th e theory of kin altruism specifi es that certain forms of prosocial behavior that are benefi cial to the recipient and costly to the helper can evolve when the benefi t B to the individual being helped is greater than the cost C to the helper, discounted by a coeffi cient of relatedness r between the helper and the individual being helped (with r = . being the degree of relatedness between identical twins, r = . between fi rst-degree relatives, r = . between grandparents and their grandchildren, or uncles and aunts and their nieces and nephews, and so on; Hamilton, ), which is equivalent to the likelihood that the recipient also possesses the helper’s “altruism gene.” In other words, specifi c forms of kin altruism are evolutionarily plausible when C < rB.
08-Baumeister_Ch-08.indd 26508-Baumeister_Ch-08.indd 265 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
core topics in social psychology
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 In support of Hamilton’s () model, people report more willingness to provide help (particularly biologically costly help) to closely related genetic relatives than to more distant ones (Bressan, ; Burnstein, Crandall, & Kitayama, ; Korchmaros & Kenny, ; Lieberman, Tooby, & Cosmides, ; Stewart-Williams, ). Estimates of migrant workers’ remittances to their families back at home based on two factors—() the fi tness costs the worker incurs by sending money back home and () the fi tness benefi ts the worker receives via the enhanced fi tness of the relatives who benefi t from the remittance—account for roughly one-third of the variance in the amounts that those workers actually send home (Bowles & Posel, ).
If humans’ penchant for prosocial behavior really did evolve in part via kin altruism, then the selection pressure for kin altruism should have left its imprint on the mind’s cognitive architecture: If ancestral humans had been unable to reli- ably identify their genetic relatives, then their prosocial behavior could not have produced benefi cial fi tness consequences for them via Hamilton’s () rule. Lieberman, Tooby, and Cosmides () outlined the workings of a hypothesized “kinship estimator” that computes the degree of relatedness between a potential benefi ciary and the benefactor. Among siblings, the kinship estimator appears to use two ancestrally reliable cues: () the degree of “maternal perinatal association” (i.e., the amount of time that the individual was in a long-term perinatal relation- ship with his or her own mother) and () the degree of sibling coresidence (i.e., the amount of time that the two individuals lived together during childhood). When these cues imply a high degree of relatedness, the benefactor is more likely to help a person in need (Lieberman et al., ). Th e challenge of identifying one’s kin seems trivial here only because we are thinking about humans—a species about which we all feel like experts rather than, say, lemurs (Charpentier, Boulet, & Drea, ), a species about which most of us know almost nothing.
Th e mind should also be sensitive to cues about the remaining reproductive potential of one’s kin because it is partly through a relative’s future reproductive potential that it is self-serving for people to provide costly help to their kin (Bowles & Posel, ). In support of this proposition, Burnstein et al. () found that participants reported more willingness to provide costly help (e.g., saving someone from a fi re) to relatives who were healthy (i.e., with greater potential for future reproduction) than to relatives who were not healthy (and whose future reproductive potential was therefore more limited).
Direct Reciprocity
Th e theory of direct reciprocity posits that mechanisms for prosocial behavior can evolve when the likelihood is greater than zero that the recipient of help
08-Baumeister_Ch-08.indd 26608-Baumeister_Ch-08.indd 266 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
Prosocial Behavior
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 will be disposed to help the benefactor in the future if the need arises (Nowak, ). Trivers () fi rst coined the term reciprocal altruism to describe this form of interaction, and demonstrated mathematically that under some condi- tions, behavioral systems for reciprocal altruism could evolve in social species.
A widely used paradigm for research on reciprocal altruism is the prisoner’s dilemma (Rapoport & Chammah, ), in which two participants are presented with a choice either to cooperate with, or to defect against, their partner. If both partners cooperate, they receive a moderate reward (the so-called “reward for mutual cooperation”). If both partners defect, both earn a small payoff called the “punishment for mutual defection.” If one individual defects and the other coop- erates, the defector receives a large boon called the “temptation to defect” and the cooperator receives the smallest payoff —the “sucker’s payoff .”
Unconditional defection is the rational course of action in the prisoner’s dilemma because it provides the best outcome both when one’s partner defects and when one’s partner cooperates. However, the prisoner’s dilemma becomes more interesting when the two individuals play multiple rounds of the game rather than only one round, allowing them to make choices based on their part- ners’ behavior in previous rounds. In a landmark study in which players from around the world submitted computer programs that would execute strategies for playing this so-called iterated prisoner’s dilemma, Axelrod () sought to determine which strategies would score the most points against all of the other strategies that were submitted.
A simple strategy called “tit-for-tat” emerged victorious. Tit-for-tat begins an iterated game with a cooperative move. If the partner also cooperates, then tit-for-tat continues to cooperate. If the partner defects on a given round, how- ever, tit-for-tat will defect on the successive round. If the defecting player ever returns to cooperating, then tit-for-tat will also return to cooperating on the next round. Tit-for-tat has several characteristics that make it eff ective in iterated games: it is () “nice” (i.e., it begins by cooperating), () retaliatory (it responds to defection with defection), () forgiving (i.e., when a defecting part- ner returns to cooperation, it returns to cooperation as well, and () clear (i.e., its decisions are honest and easy to understand). It does well in iterated games with a wide variety of strategies not by dominating them, but by racking up rela- tively high tie scores in games with other opponents that are disposed to cooper- ate and by preventing more selfi sh strategies from getting the best of itself.
Axelrod and Hamilton () demonstrated that the iterated prisoner’s dilemma provides a game-theoretic model for the evolution of Trivers’s () reciprocal altruism. Nowak () showed formally that direct reciprocity (as modeled in the prisoner’s dilemma) can favor the evolution of cooperation in social species when the probability of a successive round of interaction between two interactants exceeds the ratio of the costs of the altruistic act to the
08-Baumeister_Ch-08.indd 26708-Baumeister_Ch-08.indd 267 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
core topics in social psychology
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 benefactor divided by the value of the benefi t to the recipient. Th e fact that much, if not most, of human social life (especially the social life of small groups of hunter–gatherers and, by extension, our ancestors) involves iterated games rather than single one-shot games may explain why people in (as far as we know) every society studied to date tend to be more generous and prosocial in economic games such as the prisoner’s dilemma than standard economic theo- ries for the one-shot prisoner’s dilemma predict (Henrich et al., ; Hoff man, McCabe, & Smith, ; Simpson & Beckes, ).
Just as evolutionary psychologists interested in social behavior have deduced that the mind possesses specialized cognitive systems for computing kinship (Lieberman, Tooby, & Cosmides, ), they also have deduced that the mind possesses specialized cognitive machinery for detecting individuals who might cheat in the types of social contracts (i.e., “If you’ll help me now, I’ll help you later”) that the prisoner’s dilemma attempts to model (Cosmides, ; Cosmides & Tooby, ). Using a variant of the Wason Selection task (which illustrates that people are not very good at marshaling the right kinds of evidence to test the validity of logical statements of the form if P, then Q), Cosmides and Tooby () showed that people are more accurate at testing evidence to deter- mine whether particular individuals have cheated on a social contract. People’s relatively good skill at detecting cheaters is as true of American undergraduates as it is of people from the Shiwiar, a remote society of hunter/horticulturalists in Amazonian Ecuador (Sugiyama, Tooby, & Cosmides, ).
Others researchers have proposed that gratitude might be part of the evolved psychological system that governs reciprocal altruism (McCullough, Kilpatrick, Emmons, & Larson, ; McCullough, Kimeldorf, & Cohen, ; Trivers, ). Gratitude is a reliable emotional response to receiving help from another person that was valuable to the self, costly to the donor, and intention- ally rendered (Tesser, Gatewood, & Driver, ; Tsang, ). Th e experience of gratitude leads to reciprocation (Bartlett & DeSteno, ; Tsang, ) and strengthens relationships between benefactors and benefi ciaries (Algoe, Haidt, & Gable, ).
Forgiveness might also be an important component of the evolved psycho- logical apparatus that facilitates reciprocal altruism, and perhaps also kin altru- ism as well (McCullough, ; McCullough, Kurzban, & Tabak, ). In the context of reciprocal altruism in particular, responding to defections by occasion- ally forgiving them rather than retaliating can help to preserve cooperation when there is a possibility that individuals might make mistakes in implementing their prosocial intentions, or might mistake their partners’ prosocial intentions for selfi sh ones (Van Lange, Ouwerkerk, & Tazelaar, ). People who forgive their relationship partners for interpersonal transgressions experience greater restora- tions of positive relations (Karremans & Van Lange, ; Tsang, McCullough, &
08-Baumeister_Ch-08.indd 26808-Baumeister_Ch-08.indd 268 3/12/2010 12:38:54 PM3/12/2010 12:38:54 PM
OUP UNCORRECTED PROOF – FIRST-PROOF, 12/03/2010, GLYPH
Prosocial Behavior
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 Fincham, ) and elicit prosocial behavior from partners who have trans- gressed (Kelln & Ellard, ; Wallace, Exline, & Baumeister, ). In support of the contention that the capacity to forgive was naturally selected on the basis of selection pressure for the maintenance of valuable relationships, people are more forgiving of relationships in which the transgressor and victim are close and com- mitted (Finkel, Rusbult, Kumashiro, & Hannon, ; McCullough et al., ) and in which transgressors have communicated (e.g., through apologies or other expressions of remorse) their inability or unwillingness to harm the victim in the future (McCullough, Worthington, & Rachal, ), though these eff ects are more diffi cult to demonstrate experimentally between strangers in laboratory settings (Lount, Zhong, & Murnighan, ; Risen & Gilovich, ).
At the neural level, mutual cooperation during prisoner’s dilemmas is supported by brain regions involved in motivating the pursuit of reward (e.g., nucleus accumbens, caudate nucleus, ventromedial frontal/orbitofrontal cor- tex, and rostral anterior cingulate cortex; Rilling et al., ; Rilling, Sanfey, Aronson, Nystrom, & Cohen, ), and is partially dependent on serotonin (Wood, Rilling, Sanfey, Bhagwagar, & Rogers, ).
Social psychologists have identifi ed several other factors that infl uence cooperation in prisoner’s dilemma-type situations. For example, the ability to communicate (and, therefore, coordinate) with an interaction partner fosters cooperation in prisoner’s dilemma-like contexts (Kiesler, Sproull, & Waters, ; Steinfatt, ), especially when people can make mistakes in imple- menting their prosocial intentions (Tazelaar, Van Lange, & Ouwerkerk, ). In addition, people cooperate more with ingroup members than with outgroup members when sharing limited resources (Van Vugt, Snyder, Tyler, & Biel, ), perhaps because ingroup members are seen as more trusting than outgroup members (Turner, Oakes, Reicher, & Wetherell, ).
Indirect Reciprocity
Th e evolution of direct reciprocity requires a relatively high probability of future interactions among individuals who are taking turns helping each other, but there is a kind of prosocial behavior that can evolve even when the benefactor and benefi ciary have zero likelihood of meeting again. Indirect reciprocity occurs when a benefactor acquires a good reputation for providing help to peo- ple in need; this encourages other individuals to help the benefactor in the future (Nowak, ). Experimental evidence shows that people tend to help, donate, or cooperate mor
