Pergamon Behav. Res. Ther. Vol. 32, No. 2, pp. 237-241. 1994 Cmvrinht @3 1994 Elsevier Science Ltd Printed’k zyxwvutsrqponmlkjihgfedcb &a;Britain. All rights reserved OOOS-7967/94 $6.00 + 0.00 Subjective symptoms and cardiac reactivity to brief hyperventilation in individuals with high anxiety sensitivity GORDON J. G. ASMUNDSON,‘.~* G. RON NORTON,~ KEITH G. WILSON~ and L~RNA S. SANDLER’ ‘Department of Psychology, University of Manitoba, Winnipeg, Manitoba R3T 2N2, ‘St Boniface General Hospital, Anxiety Disorders Research Program, 351 Tache Avenue, Winnipeg, Manitoba RZH 2A6, ‘Department of Psychology, University of Winnipeg, Winnipeg, Manitoba R3B 2E9 and 4Department of Psychology, The Rehabilitation Centre, Ottawa, Ontario, Canada (Received 27 January 1993) Suntmary~ognitive models maintain that panic attacks may be initiated by fear resulting from the interpretation of somatic sensations as personally threatening or harmful. Similarly, several researchers have proposed that the enhanced response of panickers to biological challenge may result from the fear of induced anxiety sensations rather than from direct stimulation of aberrant biochemical systems. The present study examined the effects of both panic history and fear of anxiety sensations on subjective and cardiac responses to biological challenge. Eighty nonclinical subjects were chosen on the basis of level of anxiety sensitivity and history of previous panic attacks. High and low anxiety-sensitive panickers and nonpanickers (four groups of 20 subjects) were subjected to a 90 set period of voluntary hyperventilation, during which heart rate was assessed. Regardless of panic history, total symptom scores did not differ between high and low anxiety-sensitive subjects at baseline or pre-hyperventilation, but did differ at post-hyperventilation. There were, however, no significant differences in post-hyperventilation measures of heart rate. The apparent mismatch of subjective and physiological responsivity to the challenge in high anxiety-sensitive individuals (i.e. more severe symptom self-reports in the absence of increased cardiac activation) provides support for the hypothesis that high anxiety sensitivity is associated with an enhanced tendency to panic in response to biological challenge. INTRODUCTION Panic attacks are characterized by the sudden and pronounced occurence of anxiety. This anxiety is often accompanied by a wide array of physical symptoms and catastrophic cognitions. For panic-prone individuals, these symptoms occur during spontaneous panic attacks, situational panic attacks, and during the course of biological challenges such as sodium lactate administration and hyperventilation (Barlow, 1988). Panickers appear to be intensely aware of the changes within their bodies (Chambless, Caputo, Bright & Gallagher, 1984; Hibbert, 19841 King, Margraf, Ehlers & Maddock, 1986) and they often report having unusual fear regarding th&e changes (Chambless & Gracely, 1989, Clark, Salkovskis, Gelder. Koehler, Martin, Anastasiades, Hackman. Middleton & Jeavonne, 1988). Indeed, patients with.panic disorder @pee, Antis & Barlow, 1988), agoraphobia (McNaily & Lorenz, 1987) and individuals suffering from nonclinical panic attacks (Asmundson & Norton, 1993; Cox, Endler, Norton & Swinson, 1991) have all been found to score higher than nonpanickers on measures of fear of anxiety symptoms (i.e. anxiety sensitivity; Peterson & Reiss, 1987). Several researchers have suggested that the enhanced response of panickers to biological challenges (and possibly to physical symptoms occuring during spontaneous attacks) may be the result of high anxiety sensitivity. That is, panic episodes may be initiated by fear resulting from misinterpretation of the induced sensations rather than from direct stimulation of aberrant biochemical systems (Clark, 1986). McNally and colleagues (Holloway & McNally, 1987; Donnell & McNally, 1989) have investigated this hypothesis using a voluntary hyperventilation procedure. In these studies, nonclinical Ss with high anxiety sensitivity reported more physical sensations and more anxiety following hyperventilation than did low anxiety-sensitive Ss. Moreover, it was found that anxious responses were not associated with a history of panic attacks without coexisting high anxiety sensitivity. In general, the results of these investigations suggest that high anxiety sensitivity may be a necessary precondition for panic-responses to be induced by biological challenges. That is, high anxiety sensitivity may be associated with the tendency to misinterpret, in a ‘catastrophic’ way, the bodily sensations induced by such challenges (cf. Clark. 1988: McNally. 1990). One limitation of these studies, however, has b&n the lack of measurement of actus physiological change provoked by the challenge of hyperventilation. Consequently, the relationship between subjective and physiological responses in high and low anxiety-sensitive individuals remains to be determined. The present study was designed to investigate the effects of anxiety sensitivity and prior history of panic attacks on both subjective and cardiac responses to hyperventilation. In this respect, groups of nonclinical individuals with high and low anxiety sensitivity, each subdivided into groups of panickers and nonpanickers, were exposed to brief voluntary hyperventilation. Self-report measures of body sensations were administered during baseline, me-. and oost-challenae. Additionally, heart rate- was monitored during each of these phases. Given the results of -previous investigations-of hyperventilation and anxiety sensitivity (Holloway & McNally, 1987; Donnell 8c McNally, 1989) and studies of physiological responses to brief voluntary hyperventilation in medical patient populations (e.g. Beck, Berisford & Taegtmeyer, 1991), it *To whom correspondence should be addressed. 237