Abstract

Introduction

Materials & Methods

Results

Discussion & Conclusion

References

Discussion
Board

Validation criteria

In the literature on animal models of AD/HD, there has not been any systematic discussion on how to decide the validity of the model. Ideally, an animal model of a CNS disorder like AD/HD should be quite similar to the disorder in terms of etiology, biochemistry, symptomatology and treatment [30]. Thus, the most adequate model is the one that (a) best mimics, although in a simpler form than the full-blown clinical case, the fundamentals of the behavioral characteristics of AD/HD (face validity), (b) conforms with a theoretical rationale for AD/HD (construct validity), and (c) is able to predict aspects of AD/HD behavior, genetics and neurobiology previously uncharted in the clinics (predictive validity)(cf., [31; 32]).

In order to discuss Willner's validation criteria [33] as well as Taylor's [1] requirement of homology of symptoms, AD/HD symptoms will be compared with SHR behavioral deviations applying a task where AD/HD children, SHR and controls have been tested with homologous behavioral paradigms.

Both AD/HD and control children [34] as well as SHR and a large number of other strains [5; 12; 13; 19; 21] have been tested with the very same multiple fixed-interval/extinction schedule of reinforcement (Fig. 1).

All experiments were run with the same computer programs, with same type of interface, but of course with different manipulanda and reinforcers. A schedule is termed multiple when two or more schedule components operate in alternation, each in the presence of a different stimulus. The fixed-interval component measures reactivity to reinforcers, activity and motor impulsiveness. The extinction component measures sensitivity to stimulus change and sustained attention.

Back to the top.