CAREER: A neurocomputational approach to contextually flexible human social behavior

Humans often make guesses about what other people are like or what other people might do when making decisions, including decisions about when and whether to support, listen to, punish, or avoid particular individuals. In contrast to following simple decision rules across all contexts, these guesses can support flexibility in people’s decision-making, allowing them to use cues in the environment to tune their behavior to interactions with different people.

At the same time, when whole collections of decision-makers use similar cues to make similar guesses about others, this can perpetuate disparities in treatment across different social groups. This project aims to generate new scientific understanding of flexible human social decision-making, with implications for understanding social disparities. The plan is to combine methods from psychology, neuroscience, and behavioral economics to generate predictions of complex social behaviors and to test hypotheses about the neural circuitry that underlies them.

This project aims to take a step toward a comprehensive framework that incorporates more of the richness and complexity of the social world into quantitative models of human behavior, advancing efforts in the fields of behavioral economics, social cognition, and decision neuroscience. In turn, it has the potential to generate new understanding of the origins of disparate treatment.

Additional goals are to: 1) support the development of a diverse STEM workforce through initiatives aiming to lower barriers to success in the mind and brain sciences, 2) increase public engagement with neuroscience research, and 3) train students in computational modeling and cognitive neuroscience, as well as broadly appliable data science techniques.

The broad aims of this project are to: 1) incorporate more of the psychological richness of the human social world into formal computational models of behavior, and 2) use those models to test hypotheses about the cognitive mechanisms and neural circuitry that give rise to social decisions. More specifically, the research plan includes studies that integrate information about social perception (how people see others) into computational models from behavioral economics and cognitive psychology that make predictions about how people value others’ outcomes (social valuation), value outcomes at different times (temporal discounting), learn about rewards (reinforcement learning), and value information (information seeking).

The plan is to use information from these computational models with fMRI to better understand the neural basis of decisions in social contexts, with a focus on brain regions involved in social cognition and valuation. These research efforts synergize with educational and outreach initiatives to develop equitable assessment tools, engage the local community on neuroscience topics, and provide students with science mentorship, professional development, and training in broadly applicable statistical and computational modeling techniques.

The research plan aims to contribute to a more widely applicable framework capable of making better predictions of human behavior in social contexts, and the results of the research should increase our understanding of specific social decisions, how people value others, and the underlying cognitive and neural mechanisms.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.