Abstract
Background: Outcome probability bias—in the context of social anxiety—is the tendency to overestimate the likelihood of negative evaluation and contributes to the etiology and maintenance of symptoms. Research shows that reductions in outcome probability bias may be key for therapeutic improvement, yet it has historically only been measured as a unitary construct. Assessments have nearly exclusively involved self-report questionnaires consisting of verbal descriptions of social situations. The recently developed Outcome Probability Task uses pictures of social situations to assess outcome probability bias. An exploratory factor analysis previously suggested three factors: Performance situations, social gatherings, and probable judgment based on facial expressions. These factors may reflect the dimensionality of outcome probability bias, with potential to enhance assessment and treatment practices. The goal of this research was to evaluate the stability of a three-factor structure and compare it to alternative models within a new sample.
Methods: Participants (N = 283; Mage = 21.06, SD = 6.08) were predominantly students, who reported diverse sociodemographic identities and psychological characteristics. They completed the ePrime-administered Outcome Probability Task, which prompts respondents to imagine they are about to encounter the social situation shown, and rate the likelihood of being negatively judged.
Results: Fit indices and factor loadings from confirmatory factor analyses indicated that the hypothesized three-factor model is superior to the alternative models tested. Fit metrics were as follows: χ2/df = 2.34, (p < .001); Comparative Fit Index = 0.92; Tucker-Lewis Index = 0.91; Root Mean Square Error of Approximation = 0.07 (p < .001); Standardized Root Mean Squared Residual = 0.06. Factor loading estimates were acceptable (≥ 0.7) and ranged from 0.59 to 0.89 (p < .001). Covariance estimates between factors ranged from 0.55 to 0.73 (p < .001).
Conclusions: The Outcome Probability Task can examine subtypes of probability bias that may reflect differential anticipatory social fears. Findings provide additional support that the Outcome Probability Task is a tool that can be used to engage in the multi-method assessment of social outcome probability bias subtypes—which could facilitate tailored treatment approaches.