Date of Graduation


Document Type



To examine the relation between the intensity of systemic bodily events and content-specific fear conditioning to stimuli that are fear-relevant or prepared for such associations, the present study involved a direct comparison between the conditioning effects produced by 20% and 13% CO{dollar}\\sb2{dollar}-enriched air (UCSs) to three CS types: external fear-relevant (snake moving toward viewer), internal fear-irrelevant (human heart beating arrhythmically), or neutral control (flowers swaying in the wind). Sex-balanced groups of undergraduate participants (N = 96) were assigned to one of the three CSs, and within each CS, to either 20% or 13% CO{dollar}\\sb2{dollar}-enriched air (UCSs). A variety of autonomic (e.g., skin conductance responses, heart rate, frontalis EMG) and self-report measures were assessed before (e.g., Anxiety Sensitivity, Bodily Fears, Cardiac Anxiety), during (SUDS and Evaluative Ratings), and following (DSM-IV panic symptoms) three consecutive experimental phases: habituation (four nonreinforced trials), acquisition (eight reinforced CS-CO{dollar}\\sb2{dollar} pairings), and extinction (10 nonreinforced trials). Consistent with previous studies, both electrodermal and cardiac conditioned responses to fear-relevant stimuli were more readily acquired, of higher magnitude, and were more resistant to extinction compared to fear-irrelevant stimuli. This effect was especially pronounced to the heart CS and suggests a certain "belongingness" with largely cardioceptive sensations produced by the CO{dollar}\\sb2{dollar} manipulation. Frontalis EMG was not a sensitive indicator of conditioning. Self-report measures taken during the procedure also seemed less sensitive to conditioning and appeared to reflect transient state dependent effects of the UCS during acquisition. Reports of panic sensations varied reliably as a function of the intensity of the UCS/UCR, with females reporting more overall distress and panic symptoms than males. Generally, the findings suggest that the intensity of the UCR has a modulating influence in producing conditioning to stimuli that are fear-relevant or prepared for such associations. The implications of these findings are discussed in terms of current theory suggesting that panic and alarms can function as conditioning events in clinical fear onset.