The recent and pending introduction of new antipsychotic medications carries the hope of a significant advance in the treatment of schizophrenia. Although the propensity of these agents to cause fewer motor side-effects than conventional neuroleptics may lead to improved compliance and clinical effectiveness, the promise of a significant impact upon the lives of patients may primarily reside in the evidence that the atypicals alleviate negative features such as emotional flattening, social withdrawal and impoverished speech. Auditory hallucinations and delusional thinking are the more dramatic expressions of illness, but these negative symptoms, along with neuropsychological deficits, are arguably more responsible for the persisting debilitation exhibited by schizophrenics (McKay, 1980; Pogue-Geile & Harrow, 1985; Breier et al. 1991; Crow, 1991; Mukherjee et al. 1991; Kane & Freeman, 1994; Perlick et al. 1992; Green, 1996; Green et al. 1997). Negative symptoms and neuropsychological deficits are minimally responsive to conventional neuroleptics (Goldberg et al. 1991; Meltzer, 1992; Lee et al. 1994; Meltzer et al. 1994), leaving schizophrenics ill-equipped to deal with the demands of normal living.

The claim is often made that clozapine alleviates both negative symptoms and neurocognitive deficits (e.g. Meltzer, 1995a). Although there is hope that the newer antipsychotics will do likewise, the evidence for neurocognitive gains in particular is, so far, limited. Only a few studies of the effects of novel antipsychotics (such as risperidone, olanzapine, sertindole and related in-trial agents) on neuropsychological functioning have been undertaken. When effects have been demonstrated, their significance has remained unclear.

This state of affairs is unsatisfactory, as a positive impact upon neuropsychological functioning would be of interest for more than just clinical reasons. An amelioration of cognitive deficiencies would suggest that these features are not inexorably tied to an irreversible pathology, such as gross neurodevelopmental aberrations or loss of neural tissue. Rather, such gains would suggest a treatable underlying pathophysiology, lending hope to other treatments, including cognitive rehabilitation. Since these deficits are increasingly viewed as fundamental to our conceptions of severe psychiatric illness (Goldberg et al. 1991; Green, 1996; Nuechterlein & Subotnik, 1996), neurocognitive changes might reciprocally shed light on these medications and schizophrenia. Finally, differential effects on cognition across medications should be factored into cost–benefit analyses, particularly when these effects are accompanied by broader adaptive functioning gains.

Is there any reason to believe that the novel antipsychotics will significantly improve the functional capacities of schizophrenics? Several considerations are relevant, including purported action mechanisms, animal behaviour findings, neurological effects, negative symptoms effects and existing cognitive outcome data.