Comment on “Rule-governed behavior and self-control in children with ADHD: A theoretical interpretation”.

Abstract

In recent years, several papers (including the present paper) have considered phenomena putatively relevant to rule-governed behavior (RGB). At the request of the editor, I have condensed a few of my concerns about the practice of drawing casual connections between noncompliance, failure to complete assigned tasks, etc., and basic research on RGB. These are not particularly original or new concerns, since they largely echo earlier cautions about this practice (see Schlinger, 1990; 1993; 1995). Nevertheless, the central thesis of this commentary is to reassert (see Schlinger, 1990) that if RGB is to be invoked, we should reserve its use for circumstances that merit such special usage, and in particular, when direct contingency analyses no longer seem capable of accounting for or describing the phenomena of interest in a way that leads to effective action. To get directly to the point, the fact that children with ADHD have language and that their behavior seems unresponsive to long-term rewards, or fails to conform to expectations in various settings (schools and home), should not be taken as de facto evidence that RGB is implicated in the development and maintenance of ADHD. Yet this seems to be the basis of most arguments framing the analysis of ADHD in terms of rules, including Barkley's (1990). Indeed, models of ADHD might be conceptualized from a choice or delay discounting perspective equally profitably--if not more so (see Critchfield & Kollins, 2001). Especially relevant to the present case is the nature of the short-term/long-term choices made and the risk/reward ratios involved in making (or not making) healthy choices (see work by Rachlin, Baum, and others, as cited in Critchfield & Kollins, 2001). Notably, although Rachlin and other basic researchers in the experimental behavior analysis tradition are interested in phenomena relevant to the issues presented in this manuscript (e.g., self-control), their work is not discussed, presumably because these behavior analytic researchers work within a discovery model that is better characterized as bottom-up than top down. This is not surprising, given Skinner's (1950) preference for basic research and the notion that we should ultimately draw inferences from data rather than make inferences and attempt to fit the data to them. As a result, there are few comprehensive theories in behavior analysis (the matching law, perhaps a notable exception), and none that have specifically been applied to ADHD. Again, the fact that most children diagnosed with ADHD are verbal, by itself, does not implicate RGB as an important factor in the development of ADHD. Direct contingency shaping, modeling, and stimulus control/discrimination training, already well-developed explanatory concepts in the operant literature, might ultimately offer a much more comprehensive account of ADHD, if extended to for all of the relevant phenomena associated with the disorder (but see below regarding whether such efforts would be worthwhile). To make a convincing case for an independent analysis of verbal behavior (VB) effects, it is important to illustrate the inadequacies of other approaches. The present paper makes an effort to discuss the limitations of existing direct contingency approaches, but in my view, does not do enough systematically to clarify the distinctions between RGB and direct-contingency shaping. There is no certainty that a model based on RBG or VB will result in better intervention outcomes than models derived from direct contingency analyses; however, it is a certainty that no meaningful comparison can be made without a better conceptual understanding of RGB. A survey of theoretical papers and basic research in the behavior analytic tradition could do much to enhance the quality of research on ADHD and RGB. Some of the most important papers involving conceptual analyses of rules and RGB have been authored by Schlinger and colleagues (see Schlinger, 1990; 1993; Schlinger & Blakely, 1987). …