Are Humans Intuitively Cooperative, or Intuitively Selfish?

Are we instinctively selfish or prosocial? Do religious beliefs and practices make us more cooperative? These are perennial philosophical and scientific questions that we do not yet have a final answer to.

Clearly, both self-interest and other-regard are part of human nature. But it remains unclear whether religious believers are intuitively cooperative toward each other and whether non-believers are any different when facing a trade-off between group welfare and self-interest. We studied these questions in a recent behavioral experiment.

Remarkably, our results were in the opposite direction of the prevailing view in the literature!

What Is the Social Heuristics Hypothesis?

The prevailing view on the intuitive basis of cooperation in the literature is the Social Heuristics Hypothesis (SHH; Rand et al., 2012; Rand et al., 2014). SHH predicts that humans are intuitively cooperative. According to SHH, people internalize cooperation as a spontaneously triggered default strategy—as a heuristic—because such shortcuts save us mental effort and because being cooperative in general tends to further self-interest in repeated social interactions.

For instance, it “pays” to be nice to others whom we regularly interact with because people will tend to return the favor. Hence, regular interactions experienced in social networks, such as through regular participation in religious gatherings (e.g., Sunday Mass for Christians or Friday prayers for Muslims), should result in cooperative heuristics.

While the theory behind SHH makes perfect sense (Bear & Rand, 2016), the empirical support for it has been mixed.

Rand et al. (2012) first showed that cooperation increases under intuitive thinking in a series of experiments—finding that monetary contributions to a public good at net personal cost increases when people spontaneously make decisions (e.g., when under time pressure). This finding was replicated by the original group of researchers (e.g., Cone & Rand, 2014) as well as a few independent groups of researchers including ourselves (Isler et al., 2018), but it has also failed to replicate in numerous occasions, including a high-powered multi-lab replication study (Bouwmeester et al., 2017).

When these studies were brought together in meta-analyses, again, some supported the initial finding of intuitive cooperation (Rand, 2016, 2019) while others showed that the existing effect was very small, if it exists at all, and depended on a specific priming manipulation (e.g., the prompt to rely on one’s emotions) that is likely to suffer from methodological problems of experimenter demand (Kvarven et al., 2020).

What If Humans Are Intuitively Selfish?

While controversies surrounding SHH lingered on, an alternative theory for the intuitive basis of cooperation was proposed, predicting the exact opposite of SHH: This is the so-called Self-control Account (SCA, Martinsson et al., 2014).

In contrast to SHH, SCA claims that people are intuitively selfish, likely due to a visceral spontaneous reaction for self-protection, that people will tend to cooperate at net personal cost only to the extent they resist these selfish impulses by thinking reflectively.

Here is where our study comes in.

In an article that was recently published in Nature Human Behaviour, we experimentally compared these two contrasting theoretical models—SHH and SCA—in the context of religious cooperation.

Source: Pexels

More than 2,500 practicing Christians and atheists in the U.S. participated in our experiment online, and played a standard cooperation game called the Prisoner’s Dilemma, either with another Christian or another atheist. Participants were given some money to play the game, and they were asked how much of this money they would like to transfer to the other party. The money given to the other party is multiplied by two before the transfer. The more money one gives to the other party, the higher the total payout (i.e., “the public good”) and the less one keeps the initial amount of money (i.e., “the private good”). Hence, the money given to the other player is our measure of cooperation in a social dilemma between self-interest and other-regard. Half of the participants played this game against their religious in-groups whereas the other half played against their out-groups (e.g., Christians or atheists).

Furthermore, to see the effect of intuitive and reflected thinking on cooperation, half of the participants made their decision under 10 seconds of time pressure (designed to promote intuitive thinking), while the other half had to wait at least 20 seconds before making their decision (to allow for reflection).

Surprise, surprise!

Our results provided evidence for SCA—the opposite of the SHH dominating the literature. Specifically, in our tests with the highest statistical power, we found no evidence for intuitive cooperation. Rather, we found evidence that reflection increases cooperation!

First, we replicated the widespread tendency of group bias where Christians and atheists cooperated more with their own group members. Furthermore, our participants, especially Christians, seemed to be able to resist their selfish impulses and cooperate when they had the time to think and reflect on their decisions. This effect was independent of the religious identity of one’s partner in the game, and it remained stable when only those who correctly understood the logic of the economic game or only those who played this economic game for the first time were included in the tests.

In short, we found evidence for the intuitive selfishness and deliberated cooperation model of SCA and showed that this model is valid independent of group membership. These findings suggest that reflection can help mitigate cooperation problems between members of different religious groups and provide a new direction for future research.

We don’t dismiss the possibility that certain contexts promote intuitive cooperation as in SHH. However, we argue, in contrast to SHH, that reflection will in general tend to promote cooperation, such as cooperation with out-groups. While we may intuitively tend to be self-regarding, taking the time to reflect will likely improve social welfare through increased cooperation.