A Computational Value-Based Framework for Dynamic Emotion Expression, Suppression and Exaggeration

Psychological research emphasizes that emotion expressions serve important adaptive and communicative social functions to coordinate behavior between people. Yet an equally extensive body of research also suggests that people do not always express the emotions they feel, sometimes even suppressing them to the detriment of their own long-term social, psychological, and physical well-being. Implicit in this tension is the assumption that despite furnishing communicative benefits, expressing emotions, particularly negative emotions, sometimes incurs substantial costs to the expresser. However, researchers have only more recently begun to quantify these costs, and little is known about whether and how people consider these potential benefits and costs when expressing or suppressing their emotions. In my thesis, I propose that emotion expression constitutes a value-based decision and develop both a novel experimental paradigm and computational model to map out the dynamics of this cost-benefit analysis during real-time expressive decisions. I then leverage these tools to show that people flexibly regulate their expressions by calibrating the anticipated costs and benefits of expressing different emotions to the relevant attributes of the social context (e.g., self-interest, partner’s welfare, and overall inequality). Additionally, I subsequently demonstrate how interpersonal processes like reputational concern and its salience during these social interactions influences expressive decisions by selectively priming people to default to expressions of positive emotions like joy and driving them to use their emotion expressions to signal normative social preferences like inequality aversion more strongly. In sum, my thesis presents a novel approach to understanding emotion expression and demonstrates how such an approach can reveal the social cognitive processes that enable people to flexibly express, suppress, or even exaggerate their emotions in different social contexts. My work here lays a foundation for further research to examine not only how other contextual factors influences dynamic emotion expression, but also whether these expressions are effective in signaling their intentions to perceivers, and what perceivers infer from them.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.