Interoceptive fear conditioning and Panic Disorder II: Clarifying the role of CS-US discrimination

Despite the role afforded interoceptive fear conditioning in learning theory accounts of panic disorder, there exists little research demonstrating such learning in humans. A recent study by Acheson and colleagues (2007) was the first to successfully demonstrate interoceptive fear conditioning in humans. The authors employed a single-cue fear learning paradigm with 20% carbon dioxide-enriched air as interoceptive conditioned (CS) and unconditioned (US) stimuli. The current study had two aims: (1) to replicate the results of Acheson et al. (2007) while improving upon the methodology, and (2) to clarify the role of CS - US discrimination in fear learning and extinction. 104 subjects recruited from the University at Albany, SUNY subject pool met inclusion criteria and were semi-randomly assigned to four conditions: (1) CS-only, (2) CS-US Paired, (3) CS-US Unpaired, and (4) Discrimination. The discrimination condition was identical to the unpaired condition, with the exception of a light presented concurrently with the CS as a "discrimination stimulus". The study was successful in replicating Acheson et al. (2007) in that the paired condition displayed greater magnitude electrodermal and evaluative responding relative to the CS-only condition. The unpaired condition displayed greater magnitude of evaluative responding, but not electrodermal response, during acquisition. These results did not show the same resistance to extinction seen by Acheson et al. (2007). The addition of a discrimination cue successfully attenuated fearful responding to the unpaired contingency. Results are consistent with contemporary learning theory accounts of panic, and are discussed in regards to the role of unpredictability and interoceptive conditioning in the etiology of panic disorder.