Discovering the neural mechanisms of human social behavior could have a profound impact on public health, with the potential to shed light on complex disorders such as those involving paranoid, antisocial, or anxiety-related behaviors. By leveraging probabilistic computational models and model-based multimodal neuroirnaging experiments, our research will lay the foundation for a comprehensive multidisciplinary understanding of the computational, psychological, and neurobiological basis of social behavior in humans. Our main hypothesis is that when we are in an interactive social setting, our brain performs Bayesian inference of the intentions and cooperativeness of others using probabilistic representations. We propose to employ and extend the framework of partially observable Markov decision processes (POMDPs) to model prediction of hypothetical action outcomes, intentions of others, whether another human is cooperative or competitive, and to choose the best actions. This theoretical framework will be tested in parallel experiments in humans exploiting maximally the experimental advantages of model-based fMRI and intracranial recordings.
The specific aims are to: 1. Develop a multi-agent social POMDP model that allows other agents to be probabilistically modeled and that prescribes how optimal actions can be selected in s.ocial contexts; 2. Test the predictions of the POMDP model regarding belief inference by using simultaneous fMRI and. intracranial local field potential (LFP) recordings in patients to characterize the neural mechanisms underlying inference of another's intended actions and beliefs; 3. Test the predictions of the model with regard to learning optimal actions in social contexts using the Public Goods Game (PGG) and fMRI in humans; 4. Investigate neural population~level implementation of the POMDP model using recordings of LFPs in patients and fMRI in healthy individuals in two social hierarchy learning tasks.
Understanding the neural mechanisms of social behavior remains an important open question in neuroscience. The proposed research will provide a comprehensive multidisciplinary understanding of the basis of human social behavior. Public health implications include achieving a better understanding of human behavior in groups as well as abnormal social behaviors related to paranoia, trust, or anxiety.
