153number of loose associations under high stress, and the reactives were predicted to show greater improvement when low stress was compared with high stress. These predictions were based on a theoretical orientation that suggested that process schizophrenics show lower levels of cognitive development with idiosyncratic thought processes and, thus, would show little improvement from high to low stress, while reactives manifest a high level of cognitive development and more normal thought processes, which are interfered with or fragmented by stress. These predictions were supported by the results of the study. 2. BLEULER, E. Dementia Praecox or the Group of Schizophrenias. (Trans. by J. Zenkin). New York: International Universities Press, 1950. 3. DEWOLFE, A. S. Self-reports and case histories of schizophrenic patients: reliability and validity of Phillips scale ratings. J . clin. Psychol., 1968, 24, 415-418. 4. DEWOLFE, A. S. Cognitive structure and pathology in the associations of process and reactive schizophrenics. J. abn. Psychol., 1971, 78, 148-153. 5. DEWOLFE, A. S. Objective and projective measures and the process-reactive distinction in schizophrenia. Paper presented at the annual meeting of the American Psychological Association, Washington, September 1971. DEWOLFE, A. S. A word association measure of severity of current symptoms in schizophrenia.

In the physiological (polygraph) data, the sex of the E (ES) interacted with the sex of the patient (PS) or the patient's process-reactive status (PR) in only 1 of 16 analyses. Only in the heart-rate variability data was the ES by PS interaction significant (p less than .001). Sex of the E had an appreciable effect in two of the three measures of associative disturbance. The ES by PS interaction was significant (p less than .025) in the recall measure of instability of though processes, and the same interaction neared significance (p less than .06) in the multiword response measure of associative disturbance. It appears that the sex of the E should be considered in studies that use measures of associative disturbance and perhaps when heart-rate variability measures are used.