Previous research (Marshall, 1973) has shown that the most pronounced component of deficit on a choice reaction time task among a mixed schizophrenic sample involved response-selection processes. Other evidence has indicated that paranoids may be more deficient in this respect than nonparanoids. Hence, it was hypothesized that the former subgroup of schizophrenics would display response-selection deficit while the latter subgroup would display either less or no deficit. Response-selection processes were re-examined using the CRT paradigm with comparisons carried out among paranoid and nonparanoid schizophrenics and a group of nonschizophrenic controls. Results indicated that only the paranoid schizophrenics displayed abnormally retarded response-selection operations, the nonparanoid schizophrenics being nonsignificantly discriminable from the controls. It was suggested that past evidence of CRT response-selection deficit among mixed schizophrenics might have been attributable primarily to the performance of the paranoids, whose performance appears to be adversely affected by an increase in the number of dimensions relevant to response selection.