The development of new medications for the treatment of schizophrenia is closely tied to the concept of the disorder and its characteristic symptoms. In recent years, the symptoms have been divided into two broad categories: positive and negative. Positive symptoms tend to command clinical attention and to be treatment-responsive, which negative symptoms are more insidious and disabling, but less spectacular. Negative symptoms, which are similar to the core symptoms of schizophrenia defined by Krapaelin and Bleuler, have not received much attention until recently because of concern about reliability. However, rating scales with good reliability are now available for use in clinical and neurobiological studies. Clinical drug trials are also meeting the challenge of documenting the response to carefully rated negative symptoms. In addition, they are exploring the effects of medication on negative symptoms that are primary rather than secondary to reduced extrapyramidal side effects, or of medication that lowers levels of depression or decreases the demoralizing effects of positive symptoms. One ideal strategy for identifying the effects of medication on primary negative symptoms is to study patients with high levels of negative symptoms and low levels of positive symptoms, who are in the early stages of the illness and have been given minimal treatment with classical neuroleptics. Such samples are highly informative, but difficult to collect. Alternatively, when all confounders cannot be eliminated, the effects of medication can be explored using statistical techniques to examine covariance.