Year: 2026
Type: dissertation
Study design: Observational Study
Abstract: Social anxiety disorder (SAD) is thought to persist through self-reinforcing cycles linking threat perception, fear of evaluation, avoidance, and social behavior. This dissertation examined whether the nature of these daily processes generalize across individuals, cluster into subgroups, or vary by social anxiety severity. Two ecological momentary assessment (EMA) studies were conducted: a community sample with and without SAD (Study 1: N = 96; five prompts per day across two 14-day bursts) and an undergraduate sample (Study 2: N = 129; six prompts per day across 14 days). Momentary variables included social anxiety, avoidance motivation, and threat and safety cues (hurt feelings and positive affect in Study 1; felt judgment and authenticity in Study 2). Exploratory S-GIMME analyses found replicable autoregressive effects but no cross-variable paths at the group level and a two-subgroup solution. Model fit was adequate for 18 participants (23% of the sample). The subgroup solution was unreliable and therefore not interpreted in relation to hypotheses. Further, the linear multilevel models failed several assumptions checks, suggesting that these models provided an imperfect approximation of the EMA outcomes and were therefore interpreted cautiously. Findings are presented as preliminary and hypothesis-generating, with implications for design and analytic choices in follow-up analyses and future EMA studies of social anxiety.
Author: Birg, Jessica A.
Institution: University of South Florida
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7221488772
