IS BEHAVIORISM BECOMING A PSEUDO-SCIENCE?: POWER VERSUS SCIENTIFIC RATIONALITY IN THE ECLIPSE OF TOKEN ECONOMIES BY BIOLOGICAL PSYCHIATRY IN THE TREATMENT OF SCHIZOPHRENIA

Behavior and Social Issues, Fall 2006 by Wakefield, Jerome C

ABSTRACT:

Wyatt, Midkiff and Wong argue that biological psychiatry's power, not its scientific merits, explain token economies' eclipse by biological treatments of schizophrenia. However, these critiques of biological psychiatry, while partly true, ignore offsetting strengths and achievements as well as plausibility arguments that schizophrenia is partly biological; behavioral theory offers no cogent alternative account of etiology. Moreover, token-economy research failed to establish generalizability of changes to post-ward environments. Even Paul and Lentz's (1977) definitive research on token-economy treatment of schizophrenia failed to show generalization of changes to community life, and in fact, due to an inadvertent "natural experiment," revealed the instability of behavioral changes even after years of treatment. To preserve their belief system, behaviorists seem in danger of turning behaviorism into a pseudoscience defended by ad hoc hypotheses.

KEYWORDS: behaviorism, behavioral treatment, schizophrenia, mental disorder, history of psychology, history of psychiatry, philosophy of science, harmful dysfunction, biological causation; pharmaceutical industry; psychotropic medications

After a period of provocative research on token economies in the treatment of schizophrenia, interest in this modality waned to the point where the field now barely exists. Wyatt, Midkiff, and Wong's contention is that biological approaches came to dominate over behavioral methods in the understanding and treatment of schizophrenia for largely unscientific, social reasons such as the political positioning of the medical model and the power of the pharmaceutical industry, rationalized by a flawed research base.

However, I see little in Wyatt, Midkiff and Wong's articles to support this contention, and I think by and large the contention is a myth, albeit a potentially powerful, community-sustaining myth. One must be suspicious of such special pleading. The claim that behaviorist approaches were set aside by the scientific community in favor of biological approaches not because the scientific limitations of behaviorism itself allowed the alternative to become dominant but rather for nonscientific reasons surely deserves a high burden of proof, lest every scientific debate be turned by the loser into an interminable sociological investigation.

Like any causal hypothesis, the hypothesis that the waning of token economy treatment for schizophrenia is due to unscientific forces needs to be carefully assessed with full recognition of the evidential complexity of such claims. Such a claim must be supported not merely by showing that biological psychiatry has weaknesses-so does every theory!-but by showing also that behavioral approaches do not have equal or greater weaknesses. Just because one can identify non-scientific forces at work, that does not mean that there was not an underlying scientific logic that allowed those forces to hold sway. Such extra-scientific interests and constraints always exist and always exploit a science-based shift, and thus are always identifiable by those seeking pseudoscientific preservation of a discontinued view. The question is whether such processes replaced science rather than accompanied a reasonably legitimate process of scientific progress.

I will argue that a balanced view of the achievements of biological psychiatry and the history of research on token economies suggests that there was sufficient rational warrant for an emphasis on biological methods over behavioral intervention. Thus, the shift noted by Wyatt, Midkiff, and Wong was at least in substantial part based on a scientifically legitimate reading of the evidence. This does not mean, of course, that the balance might not change again in the future with new research.

I first comment on Wyatt, Midkiff and Wong's critiques of biological psychiatry and argue that the biological psychiatric record, while suffering from many deficiencies, is not as weak as they claim. I then review some of the history of research on token economy treatment of schizophrenia, emphasizing the problem of generalization of changes to post-treatment environments. I pay particular attention to the much-cited study by Paul and Lentz (1977), arguing that this study does not show what it is often taken to show.

WYATT AND MIDRIFF'S CRITIQUE OF BIOLOGICAL PSYCHIATRY

I can address only a sample of Wyatt and Midkiff s arguments for the weakness of biological psychiatry's research base. Surely they are right to this extent: twin studies, brain scan studies, and other such approaches often have unacknowledged methodological weaknesses, do not show as much as biological psychiatrists claim, or have not had the degree of substantive payoff that was once imagined. Granted, as in many fields, we know much less than some biological researchers have sometimes claimed.

However, Wyatt and Midkiff s assessment of biological research is equally unbalanced in not acknowledging the power of the biological evidence. Wyatt and Midkiff s critique of the use of twin studies that are used to demonstrate genetic components of mental disorders is marred by the fact that the main target of this critique, studies using separation of monozygotic twins at birth, is important but of limited value in the overall scheme of things because such separations as well as mental disorder are both very rare, so the number of instances is limited. They ignore the main strategy in such studies, namely, comparison of monozygotic versus dizygotic twin pairs and other comparisons of those with differing genetic and environmental overlap. It should be kept in mind that evidence for heritability and more generally for genetic and biological influences on schizophrenia is based on convergent evidence from a large number of sources, including not only twin studies and family studies but also, for example, higher rates of schizophrenia due to inbreeding, head trauma, and greater paternal age. Keller and Miller (in press) aptly summarize just one area of such evidence as follows: