Complementary approaches to social cognition
We combine careful characterization of behavior, longitudinal observations and neurophysiology to investigate the brain mechanisms of complex cognitive processes.
Social cognition in primates: Social cognition is unique in that it involves the interplay between individual and conspecific goals. An important goal of the lab is to investigate how these collaborations are computed at the level of individual neurons.
Human social cognition: How our brains facilitate - or hinder - our daily interactions with others is still poorly understood. Our group aims to better understand human social cognition at both the behavioral and neuronal levels.
Primate Social Interactions
Social cognition is unique in that it involves the interplay between individual and conspecific goals. An important goal of the lab is to investigate how these collaborations are computed at the level of individual neurons.
To achieve this goal, we are investing our efforts in implementing behavioral measures of naturally-occurring and task-based social interactions in the common marmoset.
People working in this effort: Ann Thompson, Shen Zhang, Tyler Short.
Online Experiments
Our current approach leverages online labor markets with novel experimental manipulations to assess social cognition. In our first paper using this methodology, we created a novel ‘Narrative Generative Survey’ in which we crafted unique brief narratives to assess distinct explicit attitudinal biases at the population level. The Narrative Generative approach has good convergent validity. It also revealed that participants show strong social contrast effects based on gender, socioeconomic status, and emotions but not for a character’s race. We also found a strong moderation by the respondents’ political leaning. Surprisingly, conservative respondents, compared to liberals, showed lower scores for happy characters. Paper.
We also employ more ‘classic’ approaches testing real-time social interactions in groups with distinct economic games. And are developing computational tools to understand the dynamic behaviors that arise in these contexts.
People working in this effort: Juliane Schmidt, Shen Zhang.
Ongoing collaborations
Value coding in primate amygdala
Together with Fabian Grabenhorst, we examined how single neurons in the amygdala encode different aspects of rewards. We looked at if and how single neurons in the amygdala signal the probability and the magnitude of a juice reward and if and how these neurons could combine these signals to arrive to more complex representations, like the expected value (the product of probability and magnitude) or its risk (the variance of the reward distribution). Interestingly, amygdala neurons signaled the risk of a reward but rarely its expected value. These signals were precise, abstract and adaptable. Preprint.
Conformity
Our colleagues at the Cognitive Neuroscience Laboratory developed a new behavioral task –very similar to a video game- to study real-time decisions in pairs. In their original study, they found that participants were more confident in their decisions when working with a partner, even though their accuracy did not improve. Pairs showed a regression to the mean in performance and confidence (in other words, bad players improved and good players worsened). We look forward to adapting this task to answer new questions. Preprint.