Panic disorder, phobias, obsessive-compulsive disorder and a series of other disorders are classified under the broader category of anxiety disorders in the DSM-IV (APA, 1994). The underlying assumption of this categorization is that these disorders share anxiety as common feature. However, comorbidity rates among these disorders are so high that their classification as different clinical entities is doubted and the limits between them are difficultly discerned (Klein & Pine, 2002). As a result, since accurate diagnosis is the basis of an effective therapeutic intervention, it is questionable whether the current classification of anxiety disorders facilitates or complicates clinical practice. The aim of this presentation is to propose a new theoretical perspective regarding the nature of anxiety disorders with important implications for treatment. According to this perspective, the basis of the etiological mechanism underlying anxiety disorders is the person’s dysfunctional beliefs which lead to the increase of stress that can manifest through a variety of clinical symptoms. Dysfunctional beliefs seem to be at the heart of other disorders as well, like stuttering (Bloodstein, 1995). Based on the above, it is proposed that it would be more functional to group anxiety and several other disorders within a new diagnostic category, labeled as ‘disorders of perception’. The inclusion of disorders within this category should be made on the basis of their common etiology, that is to say dysfunctional beliefs, and not on the basis of similarity of symptoms. The effectiveness of the use of cognitive-behavioural techniques for the modification of dysfunctional beliefs following such a diagnosis is discussed.