148 Using Game Theory in Primates to Study Interactive Social Behavior and its Neural Modulation

Abstract

Abstract INTRODUCTION A cornerstone of social interaction is the ability to anticipate each other's intentions or actions and is critically absent in individuals with disorders such as autism, schizophrenia, antisocial personality and traumatic brain injury. The neuronal basis and causal underpinnings of interactive social behavior, however, remain largely unknown. METHODS A formal framework for studying mutually beneficial decisions in animals is by iterated games such as the iterated prisoner's-dilemma (iPD) game. Here, we trained pairs of Rhesus monkeys to perform an iPD game in which they repeatedly interacted. We then used multi-neuronal recording, stimulation and neuro-chemical modulation to examine the role neural activity in the dorsal anterior cingulate cortex plays in interactive social behavior. RESULTS >We identify a new class of neurons in the primate dorsal anterior cingulate dedicated to predicting the opponent's yet unknown decisions, as well as distinct class of neurons that encoded the monkey's own current decisions independently of the other. Mixed population predictions of the other was remarkably near-optimal compared to behavioral decoders. Direct deep brain stimulation (DBS) of the cingulate selectively biased mutually beneficial interactions between the monkeys but, surprisingly, had no influence on their decisions when no net-positive outcome was possible. Finally, administration of oxytocin markedly increased sociability between the primates and was explained by a selective shift in the self-other encoding properties of the cingulate population. CONCLUSION We discover a specific group of neurons in the primate anterior cingulate that encode social predictions, demonstrate the effect on DBS in the cingulate on interactive social behavior and reveal the neural effect of oxytocin on prosocial behavior.