Toward a comprehensive functional analysis of depressive behavior: Five environmental factors and a possible sixth and seventh.

Abstract

In the 1990's, a resurgence of interest in behavioral theories of depression occurred. This resurgence derived from the establishment of clinical behavior analysis as a vibrant adjunct to applied behavior analysis (Dougher, 1993; 1994; 2000) with active theoretical and empirical investigations and treatment developments relevant to outpatient psychotherapeutic treatment of depression. In addition, a component analysis of cognitive-behavior therapy for depression showed that the behavioral component (behavioral activity scheduling was referred to as behavioral activation) was sufficient to explain recovery from initial depression (Jacobson et al., 1996) and at follow up (Gortner, Gollan, Dobson, & Jacobson, 1998). The cognitive component of treatment appeared to add little to the overall outcome. These findings renewed interest in behavioral approaches to depression treatment (Hollon, 2001; Jacobson & Gortner, 2000) and sparked development of a complete Behavioral Activation approach (Martell, Addis, & Jacobson, 2001; also see Kanter, Callaghan, Landes, Busch, & Brown, 2004). Other clinical behavior analytic approaches to depression have also been evaluated, including Acceptance and Commitment Therapy (Zettle & Hayes, 1986; Zettle & Rains, 1989) and Functional Analytic Psychotherapy (Kanter, Schildcrout, & Kohlenberg, in press; Kohlenberg, Kanter, Bolling, Parker, & Tsai, 2002). The above sets the stage for a re-analysis of behavioral theories of depressive behavior (e.g., Bolling, Kohlenberg, & Parker, 2000; Dougher & Hackbert, 1994; 2000; Ferster, 1973; Hoberman & Clarke, 1993; Kanter et al., 2004; Lewinsohn, 1974). In review of current behavioral models of depression, we found that several factors were highlighted. First, early models incorporated Skinner's (1953; 1974) stance that feelings such as depression are respondent bi-products of behavior-environment interactions. For example, failure to achieve sleep as a reinforcer for insomniacs might produce depressive symptoms. Subsequent models focused on direct functional aspects of depression. While many of the early pioneers of behavioral approaches to depression became focused on cognitive phenomena after initial investigations, this behavioral work remains an excellent starting point for current theory exploration. In particular these models emphasized depression as a function of positive reinforcement deprivatio n, either in terms of low density of positive reinforcement (Ferster, 1973) or low rates of response-contingent positive reinforcement (Hoberson & Lewinsohn, 1985; Lewinsohn, 1974, 1975). Depression was also conceptualized in terms of aversive control by these early theorists (Ferster, 1973; Grosscup & Lewishon, 1980) and in terms of loss of contingency through punishment (e.g., Seligman, 1975). In these models, deficits in behavior are a function of reinforcement deprivation or punishment, emotional behavior is seen as respondent, and additional behaviors, including thinking and feeling, may be described as adjunctive behavior (schedule induced). Later models viewed depressive behavior as operant phenomenon. As operant phenomenon, depressive behavior is directly subject to and maintained by positive and negative reinforcement (Hops, Sherman & Biglan, 1990). We will then review the theory and research on rule-governed behavior has been used to model depression both in terms of deficits in rule-governed behavior (e.g., Kanfer, 1970, 1971 as played out in Rehm, 1977, 1988) and excessive rule-governed behavior (Hayes, Strosahl, & Wilson, 1999). Finally, the possible role of establishing operations in depression will be explored. This paper will review each of these factors. THE REINFORCEMENT DEPRIVATION MODEL The simple notion that a lack of or reduction in positive reinforcement produces depression has been the foundation of most behavioral theories of depression. …