Date of Award


Publication Type

Doctoral Thesis

Degree Name





Psychology, Clinical.




This study investigated the quality or maturity of Rorschach object representation in nonpatient controls (N = 20) and in hospitalized neurotic (N = 24), borderline (N = 26), and schizophrenic (N = 26) patients. Inclusion of these groups was based on the prediction, from object relations theory, of a relationship between severity of psychopathology and object representation impairment. Dependent measures were two object representation scales developed by Krohn and Mayman (1974) and Blatt, Brenneis, Schimek, and Glick (1976). Cross-validation of discharge diagnosis by an independent diagnostician, based on patients' clinical case summaries, was the selection criterion for psychiatric subjects. Following determination of the clinical groups, the patients' Rorschach and WAIS protocols were collected retrospectively from hospital files. All Rorschach human content responses were identified, coded, and presented separately and randomly to the experimenter who scored them blindly, on both dependent measures, in counterbalanced fashion. Gunderson's Diagnostic Interview for Borderlines (DIB; 1977) was also modified and applied retrospectively to the patients' written case summaries. WAIS full scale I.Q., age, and education served to indicate group homogeneity. Hypotheses 1 and 2, which predicted significant between groups differences in mean object representation scores on both measures, respectively, were not confirmed. Further investigation of high scores revealed one significant contrast between the high scoring nonpatients and the low scoring neurotics on the Krohn and Mayman (1974) scale. This finding was thought to reflect neurotic inhibition or depression, rather than impaired object representation. None of the patient groups were differentiated on any of the developmental subcategories contained within the Blatt et al. (1976) measure. A third hypothesis of a significant relationship between the object representation measures was confirmed (r = .65, p < .0001). A final outcome of this study was the successful differentiation of the three clinical groups on 13 out of 29 modified DIB statements. In the discussion, it was concluded that the validities of both object representation measures, as applied to the Rorschach human response, remain in doubt. It was further concluded that the findings support the continued use of the DIB as a clinical and research tool and the use of the borderline category, as it is typically used in clinical settings. Recommendations and directions for future research were considered.Dept. of Psychology. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis1983 .K454. Source: Dissertation Abstracts International, Volume: 44-03, Section: B, page: 0915. Thesis (Ph.D.)--University of Windsor (Canada), 1983.