Journal of Consulting and Clinical Psychology 1997, W>l. 65, No. 6, 907-910 Copyright 1997 by the American Psychological Association, Inc. 0022-006X»7/$3.00 Introduction to the Special Section on Measuring Cognitive Products in Research and Practice David A. F. Haaga American University This article introduces a special section of the Journal of Consulting and Clinical Psychology titled "Measuring Cognitive Products in Research and Practice." The practice of cognitive behavior therapies—and research on their effectiveness and the validity of the models of psychopathology and therapy associated with these treatments—requires valid assessment of cognition. Cognitive variables can be considered at several levels; the main focus of the articles in this special section is on techniques suited to measurement of cognitive products, thoughts, and images of which people are conscious. Valid assessment of cognition is critical to clinical practice in general and to cognitive-behavior therapy (CBT) and re- search on CBT models of psychopathology and therapy in partic- ular. Just as behavioral assessment subserves behavior therapy and developments in brain imaging technology foster advances in neuroscience research, so too does the ability to evaluate CBT models require valid assessment of cognitive variables. Accordingly, the rise of cognitive approaches within behavior therapy in the 1960s (e.g., Beck, 1964; Ellis, 1962) and 1970s (e.g., Beck, 1976; Mahoney, 1974; Meichenbaum, 1977) soon gave rise to an explicit focus on cognitive assessment (Ken- dall & Hollon, 1981; Kendall & Korgeski, 1979; Merluzzi, Glass, & Genest, 1981). Tb help organize and guide research in this area, a number of authors have presented taxonomies of cognitive factors (e.g., Hollon & Kriss, 1984; Kendall & Ingram, 1989). For example, cognitive variables have been subdivided as cognitive structures (hypothesized inaccessible schemata guiding information pro- cessing), cognitive processes (means of transforming environ- mental input and inferring meaning from it), and cognitive prod- ucts (conscious thoughts and images; e.g., Segal & Shaw, 1988). Whereas there have been several excellent reviews in recent years of cognitive processing measures (Segal & Cloitre, 1993) or of cognitive assessments relevant to specific disorders (e.g., Glass & Arnkoff, 1994; Mizes & Christiano, 1995), there is a need for an updated genera] review of measures of cognitive products. Such measures are used in a wide range of studies of cognition in psychopathology (Dobson & Kendall, 1993) and therapy (Salkovskis, 1996); they are important in their own Editor's Note. Philip C. Kendall served as the action editor for this article. I am grateful to John Chamberlain for comments on an earlier version of this article and to the authors for contributing their insights and perspectives on cognitive assessment. Correspondence concerning this article should be addressed to David A. F. Haaga, Department of Psychology, American University, Asbury Building, Washington, DC 20016-8062. Electronic mail may be sent via Internet to dhaaga@american.edu. right and also as clues to cognitive processing and cognitive structures. The purpose of this special section is to update the conceptual and empirical status of the main methods for measur- ing cognitive products. Glass and Arnkoff (1997) review questionnaire measures of self-statements or internal dialogue, undoubtedly the most com- monly used formal cognitive assessment method in clinical prac- tice (more common still being informal clinical interviewing) as well as in cognitive-clinical research. There is skepticism about the value of questionnaire measures of cognition (e.g., Segal & Dobson, 1992); this skepticism is sometimes expressed without close attention to and consideration of the actual empiri- cal literature on reliability and validity of cognition question- naires, as several authors have pointed out (e.g., Abramson & Alloy, 1992; Mizes & Christiano, 1995; Robins, 1992). Ques- tionnaires not only have advantages for repeated use in clinical practice in terms of ease of administration and scoring (Schwartz, 1997) but also actually stack up well against other methods with respect to evidence of criterion-related validity and discriminant validity (Glass & Arnkoff, 1997). Discrimi- nant validity is a particularly important consideration for studies of cognition-emotion relations, lest researchers overinterpret correlations of thought measures with affect measures that are actually overlapping indicators of the same construct (Coyne & Gotlib, 1983; Smith, 1989). Glass and Arnkoff (1997) provide guidelines for future scale development in this area with respect to ensuring content validity from the outset. They also elaborate the conceptual issues in- volved in evaluating cognitive assessment measures and provide an analysis of the psychometric properties of 28 self-statement inventories that have each been used in at least three published studies. The encouraging data on psychometrics notwithstanding, there remain important conceptual concerns about the adequacy of self-statement inventories, concerns that in one way or an- other motivate most of the rest of the methods reviewed in this special section. For example, an item on an inventory may be endorsed, not because the respondent actually had that thought but because she or he agrees with the thought, had some thought generally like it, or is in a mood captured by the thought. It is 907 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.