In search of a 'common currency' circuit for social and non-social decision-making

About This Grant

Project Summary Social information influences many of our daily decisions, yet studies of the neurobiology of decision-making typically overlook social components. Importantly, to make a choice from multiple modalities, social and non- social information must, in principle, converge into a ‘common currency’. However, how these inputs are combined in decision-making circuits is largely unknown. To study how and to what extent social and non-social information are combined in decision-making circuits, we propose to develop a novel social foraging task for mice. In this task, two mice will simultaneously traverse a simple maze, making a left or right choice for water rewards. In contrast to when foraging alone, mice in the social task have additional information (i.e., the choice and reward history of the conspecific, gained through observation) that can be used to influence their decisions. In Aim 1, we will use reinforcement learning framework to model the cognitive and algorithmic process of how mice use social information, and test different learning rules by evaluating how well they fit mouse behavior. We will then use this model to quantify how overarching social relationships, such as biological sex, social familiarity, and hierarchical rank, influence the magnitude of which social information influences choice. In Aim 2, to understand how social information is integrated with other internal variables in decision-making circuits, we will record neurons in the posterior parietal cortex (PPC) and medial prefrontal cortex (mPFC), two central cites for decision-making in the mammalian brain, as mice perform both the social and non-social T-maze task. We hypothesize that social information will be more strongly encoded in mPFC, which has been previously associated with a wide variety of social behaviors, and that PPC will, in contrast, encode more general-purpose information about the moment-to-moment decision. This proposal aims to elucidate the overarching structure and organization of how decision-making circuits incorporate social information, and whether they serve as general ‘common currency’ circuits or support socially specific computations.