Pain and Fear in the Eyes: Gaze Dynamics Predicts Social Anxiety from Fear Generalisation

This study presents a systematic approach for analyzing eye movements in the context of fear generalisation and predicting Social Interaction Anxiety Scale (SIAS) scores. Leveraging principles from foraging theory, we introduce a composite Ornstein-Uhlenbeck (O-U) process as a computational model for social anxiety assessment based on eye-tracking data. Through Bayesian analysis, we infer the model parameters and identify a feature set for SIAS score prediction. The results demonstrate the effectiveness of our approach, achieving promising performance using Random Forest (RF) classification. This research offers a novel perspective on gaze analysis for social anxiety assessment and highlights the potential of gaze behaviour as a valuable modality for psychological evaluation.