A previous study (Judson & Katahn, 1960) disclosed significant differences between processreactive schizophrenics in the recall of friends' names over a 10-minute interval. The differences were greater than would have been expected from their recall of animal names and IQ scores. This was interpreted as reflecting a special restriction in interpersonal relationships in a generally impoverished relationship with the environment. No differences in associative interference or efficiency were found using Bousfield's method of analysis, with which Lester (1960) had previously detected differences amongst various diagnostic groups.The present study sought to extend the findings and employed both schizophrenic and nonschizophrenic patients. A total of 52 male subjects were rated on the Phillips scale for good and poor premorbid adjustment. In addition to the free recall of animal and friends' names, subjects also attempted to learn a new list of 60 words which contained animal names, persons' first names, vegetables, and professions. According to Lester, a learning-recall task is more sensitive to the presence of associative interference than the free-recall tasks.Both the process-reactive dimension and diagnostic category made significant independent and interacting contributions to the recall of friends' names, that is, the material with social connotations, but not to the recall of animal names. By subgroups, the rank order of recall from least to greatest, was process schizophrenics, process nonschizophrenics, reactive schizophrenics, reac-1 An extended report of this study may be obtained without charge from Martin Katahn, Department of Psychology, Vanderbilt University, Nashville, Tennessee 37203, or for a fee from the American Documentation Institute. Order Document No.