The atypical antipsychotics risperidone, olanzapine, quetiapine, ziprasidone, and aripiprazole have become first-line treatment for schizophrenia because they reduce the positive symptoms of psychosis but do not have a high incidence of extrapyramidal symptoms. However, these agents, like other antipsychotics, may take as long as 16 or more weeks to produce a response, and even with prolonged treatment are unlikely to evoke responses greater than 50% improvement in symptoms. This has led to the experimental use of high atypical antipsychotic doses, antipsychotic polypharmacy, and augmentation with other psychotropic drugs, all of which occur commonly in clinical practice. This article reviews the current evidence for these increasingly common means of treating schizophrenia and psychosis, with particular emphasis on polypharmacy and augmentation. To date, there are only two controlled studies of antipsychotic polypharmacy involving an atypical antipsychotic; the rest of the data are uncontrolled trials and case reports that describe a mixture of positive and negative findings. One multicenter, double-blind trial shows a faster onset of action when divalproex is added to risperidone or olanzapine than with antipsychotic monotherapy. A small double-blind study demonstrates efficacy when lamotrigine is added to clozapine. The rest of the data on augmentation with anticonvulsants are uncontrolled, and most report adverse effects. With the exception of divalproex, there are currently no compelling data to justify the use of antipsychotic polypharmacy or augmentation. Existing evidence suggests that the best treatments for schizophrenia and psychosis may be long-term trials of a sequence of atypical antipsychotic monotherapies at therapeutic doses.