The use of cognition-based behavior therapies in the treatment of mental or psychiatric disorders has been well documented (Mahoney, 1974; Meichenbaum, 1974). However, the extension of such therapies into the field of physical or somatic disorders has only just begun. The reasons for this are two-fold: (1) very little attention has been given to the behavioral treatment of somatic disorders in general (Gentry, 1975a, 1975b; Katz and Zlutnick, 1975; Williams and Gentry, 1977); and (2) the majority of studies that have dealt with the behavioral treatment of somatic disorders to date have either utilized an operant approach, which is basically concerned with the consequences of illness behavior as the focus of therapy, or have emphasized the physiological (tension) basis of bodily dysfunction as the primary target of behavioral intervention. Thus, a “blackbox” behavior modifier who is attempting to decrease spastic behavior in a patient with cerebral palsey (Sachs and Mayhall, 1971) by punishing such behavior with painful electric shocks is unconcerned about what the patient thinks, what his attitudes are regarding his spasms, and so forth; in short, the patient’s cognitions are irrelevant to successful treatment. Similarly, investigators who use EMG biofeedback to reduce headache intensity in patients suffering from tension headaches (Budzynski, Stoyva, and Adler, 1970) are only concerned with the patient’s cognitions if they in any way interfere with or enhance the patient’s ability to process the biofeedback information and subsequently to change bodily function in the desired direction, e.g., to relax and thus decrease the EMG activity in scalp and neck muscles. Some reviewers of operant conditioning of autonomic nervous-system responses, and their ultimate application to somatic disorders in humans, have indicated the importance of cognitive mediation (Katkin and Murray, 1968); while others (Crider, Schwartz, and Shnidman, 1969) have questioned the necessity of a cognitive mediation hypothesis for biofeedback and other behavioral strategies of altering somatic behavior, concluding that “it is currently a debatable question that cognitive activity per se produces any marked autonomic effect at all” (p. 458).