Investigating fear conditioning and its relation to anxiety symptoms using pupillometry and self-report measures

Abstract

Pavlovian fear conditioning is a type of associative learning that allows a subject to form an association between two items where there previously was no connection. Previous work suggests that fear learning is an evolutionary development for survival to avoid danger and fear conditioning paradigms can be used to explore the effects of fear learning in both animals and humans. Fear responses from the paradigms are similar to the development of anxiety responses in the real world and by conducting a statistical analysis of the data, our goal is to elucidate if there is a meaningful relationship between physiological and subjective measures for learning during fear conditioning. We used a fear conditioning paradigm with prepotent stimuli and an aversive scream to investigate fear learning in relation to self-report ratings (expectancy and affective valence ratings), responses to screening questionnaires about anxiety and depression, and physiological measures (pupillometry). Our paradigm consisted of an acquisition, extinction, recovery, and relearning phase. The acquisition and relearning phases are where the CS+ was presented with the aversive scream for a portion of the trials while the CS- was always presented without the scream. The extinction and recovery phases presented the CS+ and CS- without the scream. Our results suggest that there was not a significant correlation between GAD-7 scores and expectancy ratings, or GAD-7 scores and average pupil size over all phases of the fear conditioning task. Subjects rated higher expectancy for the CS+ condition over time while CS- expectancy ratings remained low. For average pupil size, both the CS+ and CS- decreased over time during the acquisition phase and decreased further in the extinction phase. Investigating age and gender across these experiments, we found no significant relationship between age and expectancy ratings, but there was a smaller mean expectancy rating difference between the CS+ and CS- conditions for women than men in the extinction phase, and a smaller mean difference for men than women in affective valence ratings. There was also more variability of average pupil size over a single trial for women than men. Further work will explore pupillometry beyond the acquisition phase and link to other physiological measures like galvanic skin response (GSR). Overall, this thesis indicates there is a discrepancy between self-report and physiological measures, like in expectancy ratings and average pupil size, but using a larger and more diverse demographic for future measures can give insight into fear learning and shape treatment for anxiety disorders.