Abstract
Displays of cooperation and altruism have become of interest in modern psychology. The current study aimed to replicate the Prisoner’s Dilemma task to investigate the effects of individual differences on cooperation, and to assess the extent to which reciprocal altruism is displayed. A group of 80 undergraduate participants were presented with a game mimicking the Prisoner’s Dilemma; they were required to choose whether to cooperate or compete with two opponents. Results have shown that likelihood to cooperate was not affected by the opponent’s previous cooperation; additionally, individual differences had no effect on cooperation rates. The following paper discusses possible psychological and evolutionary reasons for the given findings.
Keywords: cooperation, competition, altruism, prisoner’s dilemma, direct reciprocity, indirect reciprocity, individual differences.
Introduction
Altruistic behaviour can be defined as prosocial behaviour carried out by individuals who prioritise other’s wellbeing before their own (Gerrig & Zimbardo, 2002). Mechanisms of altruism are being studied to find why people aid others, including strangers, regardless of whether it benefits them or not (Fehr and Fischbacher, 2003; Gintis et al, 2003). Evolutionary theories have attempted to explain why such behaviour occurs; it has been suggested that reputation and the actions/presence of other individuals influence the extent to which altruism is displayed.
The theory of indirect reciprocity (IR) suggests that individuals tend to help those who help others, whereas direct reciprocity (DR) suggests that we help those who help us (Roberts, 2008). Some models assume that these individuals will not meet again, however, this is unlikely, as humans frequently socialise in groups (Roberts, 2008). Evidence from Wedekind and Milinksi (2000) shows that individuals who had been generous in the past received higher donations in the future, thus supporting DR.
Such theoretical suggestions have also been applied to animal subjects; Wilkinson, Carter, Bohn & Adams (2016) found that bats are prosocial; vampire bats cooperated with non-kin via blood sharing. This indicates that perhaps altruistic groups survive better than less cooperative groups, as suggested by the group selection hypothesis (Bowles, 2006).
Further evidence has shown that in both humans and nonhumans, helpful behaviour is more common toward relatives than toward unrelated individuals (Bowles & Posel, 2005; Krakauer, 2005). This is known as kin selection. Nonetheless, reciprocal cooperation is mostly displayed in humans. Nowak and Sigmund (2005) found that people are likely to help not only those who have helped them, but also individuals they have observed helping someone else. This willingness to cooperate and help others can be attributed to building a reputation for helpfulness; this increases other’s desire to cooperate with the individual.
Nonetheless, previous research suggests individual differences can influence displays of altruism, particularly when using the Prisoner’s Dilemma to study cooperation (Boone, De Brabander & van Witteloostuijin, 1999; Kuhlman & Marshello, 1975). Kurzban and Houser (2001) found that cooperation is influenced by the actions of other players in the previous round, suggesting that individual differences can influence one’s willingness to cooperate.
The current study aims to use a replica of the Prisoner’s Dilemma to assess the effect of DR, IR and individual differences on cooperation rates. We hypothesise that people are more likely to cooperate with someone that cooperated with them in the last round, based on evidence provided for DR. Additionally, we hypothesise that people show individual differences in how cooperative they are, depending on their opponent.
Method
Participants
An opportunistic sample of 80 participants (73 female, 7 male) was recruited for this study. All participants were undergraduate psychology students at Queen Mary University of London; ages ranged from 17-21 (M=18.23, SD=2.2). Participants were told that they were to take part in a study to gather empirical data for a research report. A short brief was given before the onset of the study and participants signed an informed consent form. Their rights to withdraw from the study at any time without question were also highlighted. Confidentiality was ensured by providing each participant with a unique identifier, which was also used as a login for the game played. The ethical procedures of this study were approved by the Queen Mary Ethics Committee
Apparatus
The experiment was carried out using a website (economics-games.com), where a replica of the ‘Prisoner’s Dilemma’ game was used to gather the data regarding cooperation. The game was played in Queen Mary’s laboratory of computers.
During the game, the participants were assigned a player number (e.g. Player 1) which was displayed. The participants could also view the number of euros they have won/lost, depending on whether they cooperated/competed. At the end of each round, participants could see their place in the scoreboard, which ranged from 1-9, and the overall amount of money of all other players.
Design
A within-subjects design was used in this study, as all participants responded in all categories. Each group of 9 players was randomised, based on the random identifiers assigned to every participant. The independent variable was whether participants chose to cooperate/compete, whilst the dependent variable was whether the frequency of cooperation/competition.
Procedure
Every participant was given the choice to either cooperate or compete with another player in their team. If both players ‘cooperated’, both received a hypothetical gain of €3. If both competed, each player received only €1. If one player cooperated, whilst the other competes, the co-operator lost €1 and the competitor gained €5. All participants played against two other players. Once the players had made their decisions, the screen displayed a ranking table, where they were shown the place they were in within their player group, and a table showing them their actions, their opponents action and how much they had lost/won.
Data Analysis
Participant’s actions (cooperation/competition) were assessed; SPSS was used to carry out a Chi-Squared test of independence to determine whether there is a relationship between individual differences and cooperation.
For the first dataset (see Table 1 and Table 2), the degree of freedom was 1 and the Chi-Square p-value was 0.341. Considering that the observed χ2 value was smaller than the critical value (3.84), we accept the null hypothesis as there is <0.05 probability that results are due to chance. This makes the correlation between the likelihood of someone cooperating with individuals who have previously cooperated with them, insignificant. Furthermore, Fisher’s Exact Test p-value of 0.430 was >0.05, thus providing further support for accepting the null hypothesis.