This article describes the methodology and results of an evaluation of family therapy training workshops. Cognitive skills were measured by written answers about a case history and intervention skills by the rating of videotaped interviews. Improvement was found on both variables, but cognitive and intervention skill levels and development were not related. These results have implications for training and job selection procedures.

In recent studies it has been inferred from performance on perceptual tasks, that paranoid schizophrenics showed extensive scanning and nonparanoid schizophrenics showed minimal scanning. Eye movements during free-search and size-judgment tasks were recorded for paranoid and nonparanoid schizophrenics, psychotic depressives, and control 5s. Medicated patients showed lower eyemovement rate and fixation rate than nonmedicated patients and controls but there were no significant differences due to schizophrenia. There was a significant interaction between patient type and medication for size-judgment means, but all such differences were reduced to nonsignfiicance by training. However, even after training, patients showed greater variability in size judgments than control Ss. The results failed to support the notion of extensive or minimal scanning behavior of schizophrenics. Errors by the psychotic patients in making size adjustments were not correlated with scanning behavior.

It was postulated that schizophrenics (high anxious Ss) have higher drive level than normals (low anxious Ss). Deductions of differential mance of schizophrenics and normals were tested using groups o 60 Ss in a paired associates (PA) learning situation involving the variables : (a) situational reinforcement (punishment and reward), (b) specificity of reinforcement (generalized and response specific) and (c) complexity of the task. It was found that schizophrenics performed better on simple PA and worse on complex PA than normals of matched learning ability. Punishments facilitated simple PA learning and disrupted complex PA learning compared with rewards, and the effect was most marked for schizophrenics.