The bulk of research in psychology and neuroscience has focused on studying processes in single individuals. However, in the real world people are rarely operating independently but rather in the context of a complex constellation of social relationships. Social interactions are a critical aspect of the human experience from romantic relationships, families, to professional relationships, and everyday transactions. Our lab is focused on understanding how: (a) information is transmitted through these interactions from communication and social-learning, (b) people make decisions that might impact others, (c) emotions are modulated and regulated by interactions, (c) cognition might be distributed across individuals. This type of work requires creating naturalistic interactions and modeling the nonlinear dynamics between individuals that ultimately shape our social interactions.

Emotions are coordinated, multi-system responses to events and situations relevant to survival and well-being. These responses ​emerge from appraisals of personal meaning that reference one’s goals, memories, internal body states, and beliefs about the world. Dysregulation of emotions is central to many brain and body-related disorders, making it of paramount importance to understand the neurobiological mechanisms that govern emotional experiences. Our lab has sought to understand some of the fundamental aspects of emotions. For example, What are emotions and how do they impact our behavior and social interactions? Can we objectively measure feelings beyond introspecting our subjective experiences? Do different people experience the same feelings? Do emotions help us make better decisions? We use a variety of techniques to measure emotions such as neuroimaging, facial expressions, and psychophysiology. We also attempt to develop mathematical models that allow us to predict how someone might be feeling in a specific context. Finally, we use multivariate machine-learning techniques to develop objective measures of subjective experiences and compare these across different feeling states and individuals.

How do we make decisions in the face of competing motivations? We have sought to understand how individuals integrate value from self-interested motivations such as financial incentives as well as psychological value from emotions, beliefs about fairness, and others’ intentions. We have developed mathematical models that allow social emotions such as guilt and anger to be formalized in economic utility functions. This has allowed us to identify brain regions (e.g., insula and anterior cingulate cortex) that correlate with these motivational signals and ultimately promote social cooperation. We have found that these other-regarding preferences may be processed in distinct regions of the brain and shift across development. For example, Children appear to have preferences for equality of outcomes, while teenagers eventually develop more sophisticated forms of reasoning about fairness that also includes notions of intentions. This shift in preferences corresponds to changes in cortical maturation in the dorsomedial​ ​prefrontal​ ​cortex​. Together this work has helped us understand how emotions, social preferences, and cortical maturation impact how we make decisions, which ultimately may inform the design of more effective policies in business, government, and healthcare.

A key aspect of social interactions is the process of building a mental representations of others in order to predict their behavior in different situations. This can be accomplished by integrating observations of an individual’s actions to create a model of who they are and what motivates their behavior. We have investigated how trustworthiness can be inferred in a simple repeated interaction in an investment game and find that people appear to use a general feedback driven reinforcement learning system to update their trustworthiness beliefs. This learning process leverages phasic firing of dopamine neurons to develop an expected outcome from direct feedback using a prediction error learning signal. We have also found that prior beliefs about a person based on their appearance or behavior in other contexts can impact this learning process such that learning is accelerated when the feedback is consistent with the prior expectations. Importantly, we do not need to always rely on direct experience to learn about a person’s trustworthiness. We can generate impressions about a person based on our experience with their friends. We can also vicariously learn through others’ experiences via gossip. This type of communication allows participants to not only form impressions about players whose behavior is directly unobservable, but can also enhance the relationship between the two communicating players, which ultimately promotes increased global cooperation​.

Most of the largest scientific advances have been accompanied by a methodological innovation which can provide new ways to measure or analyze a key signal. We believe that current measurement and analytic methods have severely limited our scientific understanding of emotions and social interactions and are highly committed to developing new experimental paradigms, measurement methods, and analytic techniques to study naturalistic interactions. For example, we have developed web applications to facilitate natural interactions and communication, equipment and software for measuring facial expressions in real interactions, and new multivariate analytic techniques to model the spatiotemporal dynamics of how these signals unfold.