Drug models have been extensively used to study the pathophysiology of schizophrenia. While they provide good insight into the neurobiology of this disorder, they have several shortcomings, which if known, help in the interpretation of results. In this paper we will discuss these shortcomings in general, and in relation to the N-methyl D-aspartate antagonist model for schizophrenia. This model has recently received a great deal of attention since both phencyclidine and the structurally related drug ketamine, produce symptoms that extend beyond psychosis per se to include other symptoms associated with schizophrenia. In fact, subanesthetic doses of ketamine in healthy individuals produce not only paranoia and perceptual alterations but also thought disorder, negative symptoms, cognitive deficits, as well as impairment on a number of electrophysiologic tests known to be abnormal in schizophrenia. These effects of ketamine will be discussed with a particular emphasis on implications for the pathophysiology and therapeutics of this disorder.