A pretest-posttest experimental design with matched controls was employed to test the hypothesis that chronic, psychotic patients would, after sustained contact with a nonprofessional therapist, show more positive changes in behavior than their socially isolated ward mates. This effect was expected to be a function of the pretreatment level of functioning of the patient and the length of his hospitalization.A sample of 100 psychotic patients who had had no contact with anyone other than hospital staff for at least 6 months were matched for age, suspected organic impairment, and pretreatment degree of withdrawal. One of each matched pair had 1 to 3 hours of contact per week with a nonprofessional therapist for 12 weeks. Nurses rated the behavior of all 100 patients before and after that 12-week period on the four scales of the Lorr Psychotic Reaction Profile.The results did not support the main hypothesis. They did, however, indicate interactive effects for nonprofessional therapist contact with the length of patient hospitalization. On the scale ofthinking disorganization, long stay patients with hospitalizations of more than 29 years reacted most positively to that contact. Similar interactive trends for thewithdrawal, paranoid belligerence, andagitated depressiondifference scores were noted, although the length of hospitalization for which the nonprofessional contact was most effective varied according to the rating scale employed. Length of stay was found to be a significant predictor of change of itself, long stay patients showing most remission of the symptoms related toagitated depression;however, in contrast, there was a trend for short stays with hospitalization up to 15 years, to show most improvement onparanoid belligerence.Pretreatment level of functioning was also found to be an important predictor of improvement, especially foragitated depression.The findings are interpreted as providing the beginnings of a rationale for the successful placement of volunteer nonprofessionals: the length of the patient's hospitalization to be the determining factor according to the area of behavior in which improvement is most desired.