You are invited to this week’s special talk organized by the Department of Psychology.
Inferring Minds in Interaction: Interpersonal Accuracy Across Communication Contexts
Date: 27.02.2026, Friday
Time: 12:30 PM
Room: A 130 (FEASS Building)
Murat Kezer, University of Oregon
Abstract: Effective social interactions depend on people inferring what others are thinking and feeling. The accuracy of these inferences can shape communication, relationships, and intergroup outcomes. My research investigates interpersonal accuracy, especially accuracy involving inferences of other people’s thoughts, feelings, and affect, and how it varies across people and communication contexts. In a recent scoping review that I conducted of published work using two primary paradigms for measuring accuracy, I identified major gaps in research comparing these paradigms and assessing accuracy across communication channels. My current work (also my dissertation) addresses these gaps by examining accuracy in judgments of others’ thoughts, feelings, and affect across video-mediated and in-person interactions, alongside people’s accuracy in inferring other people’s traits as a point of comparison. The findings show (1) that in-person settings favor trait accuracy, whereas video-mediated interactions show higher accuracy for inferences of mental states, (2) that accuracy emerges as a property of specific dyads rather than as a stable individual ability, and (3) that dyad members exhibit reciprocity in video-mediated interactions by showing correlated levels of perceiving accurately and being perceived accurately across mental states and traits. However, the current paradigms for studying interpersonal accuracy rely on labor intensive human coding. To overcome this issue, I have developed a natural language processing approach that estimates interpersonal accuracy by modeling semantic similarity between targets’ reports of their thoughts and feelings and perceivers’ inferences. We find that the automated estimates largely correspond with accuracy ratings by human coders, but they also reveal clear points of disagreement, especially when accuracy is low. These program of research has led me toward future research plans that integrate cultural and computational perspectives to examine interpersonal accuracy in socially consequential contexts.
Bio: Murat Kezer is a doctoral candidate in Psychology at the University of Oregon. His research examines how people understand others’ thoughts and feelings, and how accuracy in these inferences shapes relationships and decisions across in-person and mediated interactions. His work integrates social psychological theory with quantitative and computational methods such as natural language processing.