Stimulant effects in attention deficit hyperactivity disorder: theoretical and empirical issues.

The explanatory utility of a theory or model of ADHD or any disorder depends fundamentally on its capacity to address issues of causality. What causes a particular child to develop ADHD? What mechanisms are responsible for temporal and setting-related variations in symptom severity, and how are these mechanisms affected by pharmacological intervention? And, what processes determine whether gains in one domain will propagate across one or more others? It should be evident from the foregoing discussion that comprehensive answers to such questions are most likely to emerge through implementation of research strategies that (a) integrate biological and psychological levels of explanation, (b) permit analysis of causal hypotheses, and (c) address mechanisms involved in both etiology and mediation of treatment response. Although extant neurobiological studies of ADHD are as compelling as they are exciting, they are limited by a troubling reductionistic emphasis. The predominant animal models focus on a narrow range of behaviors that are presumed to be central to ADHD because of the topographic similarity they bear to those represented by the diagnostic criteria incorporated into the diagnostic nomenclature. These models would become increasingly valuable to the extent that future research examined the extent to which ecologically relevant behaviors (e.g., social behavior) are compromised in the animal strains and whether the observed compromises are parallel to the correlates of ADHD observed in humans. Similarly, human molecular genetic studies have provided a glimpse into the possible role that genes related to dopaminergic neurotransmission may play in the etiology of ADHD. Yet, the features of ADHD have been conceptualized in these investigations as a unitary collection of characteristics, and this has precluded analysis of what specific syndromal feature (if any single one) is affected by the implicated genes. It is intriguing to speculate whether varying combinations of genes governing properties of DA receptors and reuptake molecules are associated with different patterns of symptom severity or responses to stimulant medications. As testing procedures for determining genotypes with respect to these features become more affordable and available, it should become increasingly feasible to examine such issues empirically. Research on the utility of stimulant drugs as a treatment for ADHD also has yielded useful information. Although the effects of MPH are of short duration, the breadth of their impact is impressive. The clinical effectiveness of these medications is no longer in doubt, and patterns of relations among outcome measures represent a potentially fruitful target of scientific inquiry. Finally, data supporting a neurobiological substrate for ADHD, evidence indicating that task and setting variables moderate the expression of the syndrome's diagnostic features (see Barkley, 1998, for a review), and the causal emphasis of the conceptual model with which the discussion began collectively argue for a diathesis-stress conception of the syndrome. And, as foregoing comments make clear, task and setting variables and the mechanisms through which they influence symptom expression are as important to the phenomenon as are neurobiological predisposing causes. This has significant implications for assessment strategies employed in diagnosis and evaluation of treatment-outcome. Specifically, it suggests that theory-based experimental manipulations of task and setting variables designed to impose challenge on hypothesized core features of the disorder are more likely to yield insights into the causal mechanisms governing behavioral organization in affected children than strategies emphasizing static identification of diagnostic correlates. It is hoped that such an approach will accelerate the discovery of increasingly effective assessment and intervention strategies.