Proceedings of the Annual Meeting of the Cognitive Science Society

Cognitive search, conceptualized as information foraging through mental spaces, underpins many daily tasks and is frequently performed jointly. Yet, the mechanisms by which social interaction impacts search dynamics remain understudied. Through three agent-based simulations of a verbal fluency task, we investigated how cognitive diversity and interaction structure modulate collective performance in joint cognitive search. In Experiment 1, under strict turn-taking, moderate diversity benefited performance by inducing increased exploration, but high diversity levels proved detrimental. Experiment 2 revealed that flexible interaction protocols mitigated these costs by enabling distinct, more adaptive search strategies. Experiment 3 demonstrated that enhancing individual cognitive flexibility via working memory mitigated risks related to high diversity. Collectively, these results identify flexibility, whether situated in social interaction protocols or individual cognitive mechanisms, as a critical route for unlocking the benefits of diversity in collective search, and offer a novel computational paradigm to study the emergent mechanics of collective intelligence.