Bleuler1 cited, as one of the primary symptoms of schizophrenia, a disturbance of associative processes. Since then many studies have attempted to determine the nature of this disturbance. Early studies, Pfenniger,2 Kent and Rosanoff3 indicated that schizophrenics tended to give more idiosyncratic responses than normal controls on free association tests. Since then, free association tests have been one of the primary methods used to investigate the schizophrenic associative disturbance. Numerous studies have supported the results of early studies, ie, populations of schizophrenic Ss tend to give more idiosyncratic responses than normal controls, eg, Moran,4 Shakow,5 Sommer,6 Dodecki et al,7 and Storms.8 There are several observations which should be made concerning the interpretation of these results. Most studies showed that the great majority of schizophrenic Ss gave responses which fell within normal limits, ie, there was a