SocialAI Research Group

Modeling Rigidity or Adaptation in Shared Reality

How do broad societal-level structures and individual-level cognitions and behaviors interact to facilitate group cooperation or its downside, group conflict?

For instance, how do environmental demands (e.g., resources, geographic separation) influence the way that groups form and interact with each other which, in turn, impacts the way that individuals identify with and cooperate within those groups?

From the other direction, how do individual decisions about group identification and cooperation aggregate up into group behaviors that reshape the demands of an environment? 

This line of research explores the complex interactive processes that produce sustained group cooperation across multiple levels – the individual (individual-level), the groups with which they identify (meso-level), and the broader societal structures and norms that shape group relationships (societal-level). We test whether, and if so how, altering the societal level, such as by increasing resource scarcity, raising population densities, or introducing novel groups, requires individual agents to adapt their previously learned rules and behaviors.

The results from such simulations reveal which previous rules and norms are maladaptive in new environments, versus which norms may persist as adaptive even in new settings. Additionally, the results show how new individual or group-level behaviors may emerge dynamically as broader environments change their demands. Fundamentally, this research formalizes how norms and behaviors can evolve over time not only as a function of individual behaviors and choices but as an interactive process between the environment and the individual, with the end goal of exposing sources of inter- and intra-group conflicts.

SocialAIGroup 2022