Following relapse after 11 sessions of cognitive behavioral treatment of panic disorder, a client wore an ambulatory physiological monitoring device for 1 day. The physiological data were used to test the client's hypothesis that panic attacks always occur from “out of the blue.” Results showed that hyperventilation and fearful thoughts preceded panic attacks, refuting the client's initial hypothesis. A comparison of the client's self-report data before and after wearing the device indicated that this therapeutic strategy had a beneficial treatment effect. The results are discussed in the context of the scientist-practitioner approach to treatment failure.