This paper reviews the recent parent training research in which parents are taught to use principles of operant learning as well as general principles of positive verbal discourse. While this combination of interventions has been shown to have greater impact than the operant guidelines used alone, a rationale for the discourse strategy is not clear. In this review, it is argued that the two interventions can be understood within an expanded view of reinforcement theory. It would seem that the interventions used in concert lead to improvement in children's and parents' observational skills, as well as in their willingness to cooperate. Keywords: parent training research; operant learning; positive verbal discourse; reinforcement theory; observational skills. ********** When parents seek professional help for their children's behavior, a majority of the referral problems amount to aggression and noncompliance (Denham, Workman, Cole, Weissbrod, Kendziora, & Zahn-Waxler, 2000; Heller, Baker, Henker, & Henshaw, 1996). These two child behaviors are clearly related as a class of disruptive actions, and it is often true that noncompliance sets the stage for aggression because troubled parents are apt to provoke their children into following acts of disobedience (Patterson, 1982). Thus, a child's chronic noncompliance with parent instructions and rules seems to be a keystone part of this disruptive class, suggesting that ways of teaching children to obey their parents might also weaken the entire class of problem behaviors. Because of the above findings and assumptions, it is not surprising to see that today's parent training programs for disruptive children target both components of this response class (Barkley, 1997; Bor, Sanders, & Markie-Dodds, 2002; Webster-Stratton, 1996). All of these programs follow principles of operant learning but also make use of adjunctive strategies aimed at verbal discourse exercises, designed to promote stronger positive connections between parents and their children (see Conduct Problems Prevention Research Group, 1999; Eyberg, 1988; Feinfield & Baker, 2004; Forehand & McMahon, 1981; Schoenwald, Sheidow, Letourneau, & Liao, 2003; Webster-Stratton, Reid, & Hammond, 2004). These exercises involve friendly talk, listening, feeling talk, problem-solving discussion, art projects, and appropriate expression of feelings. In some of the above listed studies, both the operant learning and the adjunctive strategies were shown to impact referral problems of the children. Regardless of whether or not the two strategies were assessed and analyzed as independent sources of change, the authors considered both interventions to be important components of the treatment process. Based on the conceptual arguments and research reviews by Cavell (2001), Strand (2000a and 2000b), and Cavell and Strand (2002), there is reason to view these disparate intervention strategies as more similar than they seem to be. These authors' expansion of reinforcement theory includes stimulus control constructs that could place both interventions within a broader conception of reinforcement theory. In this paper I hope to further develop these ideas and, in the process, to layout some suggested guidelines for clinicians who work with the parents of disruptive children. The Holistic Nature of Parent-Child Interactions When parents are taught to train their children to obey instructions, the clinical teaching and subsequent training are often based on direct reinforcement strategies. That is, child compliance is positively reinforced and noncompliance is either ignored (extinction) or timed out (absence of opportunity to obtain positive reinforcers). This intervention step is accompanied by similar contingency management of other responses in the class of disruptive behaviors, such as demands, nagging, hitting, and property destruction. While the treatment strategy makes sense if we view each response as independent, such a perspective overlooks the research evidence documenting functional connections across these topographically different responses. …