University of British Columbia
Eyes on Me : an eye-tracking investigation of social anxiety and attentional biases on Zoom
Abstract
dc:descriptionSocial anxiety is characterized by attentional biases thought to maintain distress during social interactions, yet whether these biases translate to live videoconferencing (VC) environments remains poorly understood. VC platforms are a central medium for social, occupational, and healthcare contexts, and research suggests that unique features, such as constant availability of the self-view icon, may exacerbate social anxiety by prompting self-focused attention, a key mechanism in its maintenance. While attentional biases in social anxiety have been well-documented in face-to-face and pre-recorded contexts, few studies have examined gaze patterns during live, real-time digital interactions. The present study used dynamic eye-tracking technology to investigate the relationship between social anxiety and visual attention during a live Zoom interaction. Twenty-eight female-identifying students engaged in three social tasks with a female confederate: (1) a Neutral Autobiographical, (2) a Neutral Conversation, and (3) an Evaluative Speech task with gaze recorded across three regions: the self-video icon, the confederate’s icon, and non-social areas. Self-report measures assessed social anxiety, trait and state anxiety, self-esteem, depression, narcissism, and Zoom familiarity. Contrary to predictions, social anxiety did not predict greater self-viewing tendencies in any task. However, preliminary findings revealed an association between social anxiety and biased attention toward non-social areas of the Zoom screen during the Neutral Autobiographical and Evaluative Speech tasks, consistent with attentional avoidance. Exploratory analyses suggested that heightened post-speech anxiety predicted reduced self-viewing, suggesting self-view disengagement may serve as a virtual safety behaviour, and found that depressive symptoms predicted decreased non-social viewing during the evaluative condition. Findings suggest that social anxiety-related attentional biases in virtual spaces may be more context-dependent than current cognitive-behavioural models predict, and that VC platforms may accommodate unique attentional strategies with implications for virtually delivered mental health care.
Degree
thesis:*- Name thesis:degree_name
- Master of Arts - MA
- Level thesis:degree_level
- master's
- Discipline thesis:degree_discipline
- Psychology - Clinical Psychology
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Barg, Samantha
- Contributors dc:contributor
-
- Libben, Maya
Rights
dc:rights- Statement dc:rights
-
- Attribution-NonCommercial-NoDerivatives 4.0 International
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/94805