The severity of schizophrenic symptoms was examined in 50 male chronic patients while neuroleptic free for at least 3 weeks and during 6 weeks of treatment with haloperidol. The results suggested that 50% of the improvement associated with haloperidol administration occurred by the end of the first treatment week and that early improvement, at both 1 and 4 weeks of treatment, was predictable from drug-free symptoms severity. There was a negative correlation between week 1 improvement and improvement during the next 3 week of treatment, suggesting that medication response is not linear. Finally, dose increases after 4 weeks of treatment with 20 mg of haloperidol did not lead to any clinical improvement. These results are discussed in terms of their implications for selecting chronic schizophrenic patients who will and will not benefit from medication treatment.