Modeling Social Interactions: New Directions in Affect Control Theory

June 22-24, 2017 • Dartmouth College • Hanover, New Hampshire

Affect control theory (ACT) examines how sentiment norms – our culturally shared meanings for particular types of actors, behaviors, emotions, and social settings – organize social life and direct us toward a mutual interpretive framework for interaction. As common cultural knowledge, sentiment norms allow us to plan for, interpret, and effectively respond to social events, based on our impressions of who has done what to whom. By measuring both normative sentiments and how these sentiments shift when they combine in the context of social events, affect control theorists have built causal models of the relationship between interpretations of events and patterns of social action (Heise 1979, 2007; MacKinnon 1994). These "impression change" models were first developed in the late 1970s, and have been used in conjunction with data about cultural sentiment norms to run mathematical simulations of social interaction. Event simulations generate testable predictions about behavioral and emotional responses to social events, which have been supported by extensive survey, experimental, and naturalistic evidence in a research program spanning several decades (e.g., Robinson and Smith-Lovin 1992; Smith-Lovin and Douglass 1992; Smith-Lovin and Heise 1988).

A wave of recent methodological innovations and theoretical developments has rapidly pushed knowledge forward in this subject area (for a recent review, see MacKinnon and Robinson 2014). In the past several years, affect control theory has been expanded to account for the role of the self and social institutions in shaping situational identity dynamics. A Bayesian extension of the theory has enabled new research on the impacts of cultural diversity, situational uncertainty, and communication noise on interaction dynamics. A massive multinational data collection effort has produced new models of social interaction in U.S. culture, and enabled research on impression change processes in Arabic language cultures like Egypt, Kuwait, and Morocco. Scholars have explored new methods of estimating impression change equations and assessing the quality of their performance, and examined individual differences and change over time in the central mechanisms of impression formation. Modeling Social Interactions will review these and other new directions in affect control theory, and serve as a space for meetings on new and ongoing collaborations.


This event has been sponsored by the Neukom Institute for Computational Science, the Office of the President, the Office of the Dean of Faculty, and the Program in Quantitative Social Science at Dartmouth College.