Journal of Personality ind Social Psychology 1987, Vol. 53, No. 4,734-742 Copyright 1987 by the American Psychological Association, Inc. 0022-3514/87/100.75 Cognitive Specificity in Emotional Distress Rick E. Ingram San Diego State University Timothy W. Smith University of Utah Philip C. Kendall Temple University Christina Donnell University of Health Sciences, Chicago Medical School Kevin Rpnan Temple University Cognitive approaches to emotional distress posit that specific cognitive factors are critically linked to the etiology, course, or treatment of dysfunction. Although a number of empirical studies have assessed cognitive factors in emotional disorders such as depression and anxiety, research has yet to assess these variables simultaneously and with identical cognitive measures. Using depression and test anxiety as models of dysfunctional affective states, we examined cognitive specificity on mea- sures of information processing, attributions, automatic thinking, and cognitive interference. Re- sults indicated a pattern of specificity showing several differences and similarities in depression and anxiety. Specifically, "purely" depressed individuals showed evidence of selectively processing de- pressive information, making dysfunctional attributions, and engaging in more negative automatic thinking. "Purely" anxious individuals, on the other hand, showed evidence of selective anxious information processing and increased cognitive interference. Results are discussed in terms of a taxonomy for classifying depressive and anxious cognition. Much personality theory and research has focused on the variables underlying dysfunctional personality processes. Of special interest to many researchers in this area have been the processes linked to emotional arousal and dysfunction (e.g., Diener&Iran-Nejad, 1986; Gotlib& Meyer, 1986). In this vein, a number of investigators have suggested that social-cognitive variables may be important determinants of emotional disregu- lation. Such social-cognitive approaches are typically predi- cated on the assumption that particular cognitive factors are critically linked to the etiology, course, or treatment of mal- adaptive emotion. An abundance of theoretical work, for exam- ple, has accorded a causal role to cognitive variables in the onset of a variety of psychological disorders (e.g., Beck, 1967; Ellis & Grieger, 1977; Mahoney, 1977). Likewise, though not necessar- ily ascribing causal significance, others have maintained that cognitive variables play a fundamental role in serving to perpet- uate and possibly exacerbate emotional distress once initiated (Ingram & Smith, 1984; Smith & Greenberg, 1981). Further- more, numerous therapeutic approaches have also been devel- oped that seek to treat psychopathological emotion by specifi- cally altering some aspect of what Kendall (1985) has referred to as distorted or deficient cognitive functioning (e.g., Beck, Rush, Shaw, & Emery, 1979; Emery, Hollon, & Bedrosian, 1981; Kendall &Hollon, 1979;Meichenbaum, 1977). In line with this conceptual emphasis, empirical research has Correspondence concerning this article should be addressed to Rick E. Ingram, Doctoral Training Program In Clinical Psychology, Depart- ment of Psychology, San Diego State University, San Diego, California 92182 or to Philip C. Kendall, Department of Psychology, Weiss Hall, Temple University, Philadelphia, Pennsylvania 19122. begun to examine the cognitive factors associated with emo- tional distress. Of particular interest to many social-cognitive researchers have been the cognitive processes involved in affec- tive states such as depression and anxiety. In the area of depres- sion, for instance, investigators have found patterns of dysfunc- tional information processing (Clark & Teasdale, 1982; Derry & Kuiper, 1981; Ingram, Smith, & Brehm, 1983), an increased occurrence of irrational beliefs (Nelson, 1977), increased nega- tive self-statements (Barrel! & Ryon, 1983; Hollon & Kendall, 1980), excessive and negative self-focused attention (Ingram, Lumry, Cruet, & Sieber, 1987; Ingram & Smith, 1984; Smith & Greenberg, 1981), self-debasing attributional patterns (Ham- ilton & Abramson, 1983; Seligman, Abramson, Semmel, & von Baeyer, 1979), and learned helplessness deficits (Seligman, 1974, 1975). Similarly, research on anxiety has found dysfunc- tional variations in information processing (Mueller & Thomp- son, 1984; Smith, Ingram, & Brehm, 1983), deficits in rational beliefs (Gormally, Sipps, Raphael, Edwin, & Varvil-Weld, 1981; Himle, Thyer, & Papsdorf, 1982), patterns of anxious self-state- ments (Glass, Merluzzi, Biever, & Larsen, 1982; Smith, Hous- ton, & Zurawski, 1984), increased self-focused attention (Carver, Peterson, Follansbee, & Scheier, 1983; Sarason, 1975; Wine, 1971), dysfunctional attributions (Girodo, Dotzenroth, & Stein, 1981; Phillips, Pitcher, Worsham, & Miller, 1980), and evidence of learned helplessness (Dweck & Wortman, 1982). Despite similarities in the cognitive approaches to these emo- tional conditions, studies have typically not been concerned with comparative data pertaining to the specificity of cognitive processes in different dysfunctional states (see also Ingram & Kendall, in press). Several recent research efforts, however, have suggested the usefulness of examining such comparative data. 734