New Study Suggests Gaming Evolved To Identify Skilled Allies In Dangerous Situations

New research suggests that human gaming may have evolved as a subtle way to identify skilled allies in dangerous environments, rather than simply serving as entertainment or social bonding. The study combines laboratory experiments with computer simulations to explore the deep evolutionary origins of gaming behavior.

The findings, published in Evolution and Human Behavior, suggest that the social benefits of gaming depend heavily on context. In low-risk everyday situations, games may not create stronger bonds than other shared activities, but in high-risk settings, they could help people identify capable allies.

From Playful Behavior to Structured Games

The researchers distinguish between play and gaming, arguing that they serve different evolutionary functions.

Play is described as spontaneous, open-ended behavior without strict rules or clear goals. It is commonly observed in children and many animal species and is thought to facilitate skill learning and development.

Gaming, in contrast, involves structured, rule-based challenges that often resemble hunting, conflict, or strategic competition. Because gaming appears later in development and is considered largely unique to humans, the researchers investigated which evolutionary pressures may have favored its emergence.

Lead author Yago Lukševičius de Moraes has previously studied male psychology and social behavior. Earlier work led him to suspect that gaming contributes more to alliance formation and friendship building than to mating or parenting success, motivating the current study.

Testing Games in the Laboratory

In the first phase of the study, 40 adults aged 18 to 40 were paired into same-sex groups of strangers. Same-sex pairings were used because earlier theories suggest that men in particular may use competitive activities to build alliances.

Before interacting, participants completed questionnaires evaluating their partner’s perceived qualities, including intelligence, cooperativeness, and charisma. They also rated relational proximity, a measure of perceived emotional closeness.

Half of the pairs played Nine Men’s Morris, an ancient strategy board game involving placement and movement of pieces to capture an opponent’s tokens. The remaining participants completed a role-playing exercise in which they acted as friends with conflicting goals who needed to negotiate an agreement.

After the activities, participants repeated the questionnaires and later returned to the laboratory 14 and 28 days afterward. During subsequent visits, activities were switched so that all pairs experienced both gaming and role-play conditions.

The researchers predicted that gaming would accelerate social bonding, leading participants to stabilize judgments about each other’s value and experience stronger feelings of closeness compared with role-playing tasks.

However, the results did not support a unique social effect of gaming in this low-stakes setting.

Changes in perceived partner value and relational closeness were primarily explained by repeated exposure over time. Simply spending time together, regardless of whether participants played games or completed conversational tasks, appeared to drive the observed social changes.

Simulations of Ancestral High-Risk Environments

To explore scenarios beyond laboratory limitations, the researchers used agent-based evolutionary modeling.

In these computer simulations, artificial individuals followed simple behavioral rules, allowing researchers to observe how traits spread across many simulated generations.

Each population began with 100 agents, including 99 non-gamers and one mutant gamer. During each generation, agents selected one cooperative partner to complete a risky task resembling ancestral hunting or warfare scenarios.

Survival probability depended on the combined skill levels of the paired agents and an environmental hazard parameter representing the consequences of failure. Successful agents reproduced and passed on their traits, whereas unsuccessful agents were removed from the population.

Gaming behavior spread only under certain conditions. Most importantly, the original gamer needed to possess above-average skills so that others benefited from interacting with them. When the initial gamer had poor skills, the gaming trait consistently disappeared from the simulated population.

One unexpected finding was that gaming could still evolve even when agents did not remember previous partners. This suggests that the social advantages of gaming may emerge from brief, one-time interactions without requiring long-term relationships or strong memory of prior allies.

Risk Levels and Skilled Allies

The simulations also showed that environmental danger needed to increase alongside overall population skill levels for gaming behavior to remain advantageous.

As populations became more competent, hazards also needed to become more severe in order for differences in skill to remain meaningful.

In relatively safe and stable environments, the advantage of identifying slightly more capable partners diminished, causing gaming behavior to gradually disappear. In dangerous environments, however, identifying a highly skilled ally could become critically important for survival.

Together, these findings support what the researchers describe as the competition-for-allies hypothesis. According to this framework, gaming helps individuals signal and identify valuable partners in environments where skilled allies are relatively rare and tasks involve substantial risk.

Co-author de Moraes notes that although many players report feeling more connected after gaming, social bonding is not guaranteed. Its effects appear to depend on factors such as game type, relative player skill, and broader social and environmental context.

Limitations and Future Research

The authors emphasize that both the experimental and simulation components represent early stages of investigation.

The laboratory study involved a relatively small sample size and only one type of board game, limiting generalizability across cultures, age groups, and gaming formats.

Future studies may involve larger and more diverse samples, longer interaction periods, and a broader range of activities, including sports, video games, tabletop role-playing games, and games based primarily on chance. More standardized measures of social bonding could also improve comparisons across studies.

On the modeling side, the researchers plan to incorporate additional mechanisms such as social learning, where agents imitate successful strategies, and genetic relatedness, which may influence alliance formation dynamics. These additions could more closely resemble real human social systems.

Another unresolved question is whether gaming may function as a low-cost mechanism for conflict resolution. Strategic games may provide structured environments in which status, competence, and hierarchy can be negotiated without physical aggression.

Because gaming is widespread yet often studied in fragmented ways, the authors call for greater interdisciplinary collaboration between psychology, anthropology, game studies, and evolutionary biology.