Nondirective (ND), applied relaxation (AR), and cognitive behavioral (CBT) therapies for generalized anxiety disorder (GAD) were compared. The latter 2 conditions were generally equivalent in outcome but superior to ND at postassessment. The 3 conditions did not differ on several process measures, and ND created the greatest depth of emotional processing. Follow-up results indicated losses in gains in ND, maintained gains in the other 2 conditions, especially CBT, and highest endstate functioning for CBT. AR and CBT thus contain active ingredients in the treatment of GAD; support exists for further development of imagery exposure methods or cognitive therapy because of their likely role in promoting maintenance of change with this disorder. Expectancy for improvement was also associated with outcome, suggesting the need for further research on this construct for understanding the nature of GAD and its amelioration.Controlled studies of cognitive-behavioral treatments (including relaxation, anxiety management, and cognitive therapy) for generalized anxiety disorder (GAD) have found these techniques to produce greater improvement than no treatment and to yield maintained gains up to 2 years later, despite chronicities of several years (Barlow et al, These initial studies have been disappointing, however, in two ways: The degree of clinically significant change has not been large, and comparisons among various therapies or elements have been generally nonsignificant. Thus, research has yet to rule out rival hypotheses concerning the active ingredients of treatments for GAD, including the crucial possibility that nonspecific factors account for the observed gains.In the present study, we attempted to increase the efficacy of a behavioral and a cognitive-behavioral treatment by certain technique emphases based on theory and recent basic research findings regarding the nature of GAD. Moreover, an attempt was made to conduct a controlled outcome investigation with a methodology and design (contrasting applied relaxation [AR], cognitive-behavioral therapy [CBT], and nondirective therapy [ND]) that would hopefully allow unambiguous elimination of two of the most likely rival hypotheses in prior studies: nonspecific factors and the relaxation training component.As does all anxious experience (Lang, 1968), the anxiety of a GAD client involves a process of interacting systems (attentional, conceptual, imaginal, physiological, affective, and behavioral) that unfolds over time in continual response to a constantly changing environment. Several features, however, may distinguish the process of GAD from those of other anxiety disorders (cf. Borkovec, Shadick, & Hopkins, 1991): (a) less obvious environmental triggers of the anxiety or behavioral avoidance to them; (b) preattentive bias to multiple threat cues with rapid cognitive avoidance on detection; (c) predominance of verbal-linguistic worry, which may function to avoid aspects of anxious experience by suppressing fear-related imagery, somatic activation, and emotional proce...