New Yale research hopes to bring society one step closer to understanding how we can all just get along.

For over 30 years, psychologists have examined why individuals cooperate under specific circumstances, but little work has been done to examine if there are patterns of cooperation that appear consistently across contexts and that might be caused by an underlying cognitive characteristic. By studying common traits in how subjects responded to a range of computerized decision-making games, the team of researchers from Yale and Harvard identified patterns in cooperative behavior suggesting that such an trait does exist. These findings debunk assumptions about cooperative behavior that have guided research in the past and raise new questions regarding the nature of why people pay costs to benefit others, said David Rand, a study senior author and professor of psychology, economics and management.

“We found substantial evidence that how someone acts under one set of conditions is highly predictive of how they will act under other circumstances, which, in turn, provides strong evidence that cooperativeness is an element of the person,” he said.

To investigate cooperative behavior, Rand and his team used a series of computer game simulations, based on game theory models, that offer each participant the choice of whether to help themselves or others. A major class of simulations were economic games, which endowed each participant with a certain amount of money that they could then decide to keep or use to benefit others or a collective group effort. In one of these simulations, players had the option of contributing their money to a collective pot, which was then doubled and split between members. Also included was a version of the prisoner’s dilemma, where subjects had the opportunity to sacrifice funds to punish another person who was acting selfishly.

The research team found that how a participant behaved in one game was predictive of how they would behave in another cooperative situation, which Rand said suggests a common source of cooperation.

To measure how likely respondents were to make the same game decisions on separate occasions, the study analyzed the behavior of those who participated in the games more than once, an average of 124 days apart. Responses were found to be consistent over time.

Moshe Hoffman, a research scientist at Harvard’s Program for Evolutionary Dynamics who was not involved in the study, said that the research is important because it shows that measures used in cooperative behavior studies reflect deeper cognitive traits and thus have applicability outside academia.

Jillian Jordan GRD ’19, a graduate student working with Rand at Yale who was not an author of the study, said that future research in the field will further examine the predictability of cooperation in various settings.

“There’s an assumption of certain reputation-based theories that people cooperate to signal that they are ‘nice guys’ who will cooperate in the future, or in other situations,” Jordan said. “This idea seems intuitive, but it doesn’t have a ton of empirical support in the literature, and it’s very important to know if economic cooperation games can tap into this ‘nice guy’ concept.”

Though the study offers evidence that a cooperative core trait exists, Rand said the study does not address how much of that phenotype is due to an individual’s genotype and how much is due to environmental factors.

The study appears today in the journal Nature Communications.