The quality of object relations and nature of attachment in homosexual and heterosexual men

The contention underlying this investigation is that homosexuality continues to be viewed by clinicians and theorists, as well as laypeople, as psychopathological due to confounding relationship pathology with the sexual orientation of clinical cases who happen to be homosexual. It was predicted that degree of relational commitment and satisfaction, and not sexual orientation, would account for any observed/assessed object relations or attachment pathology in homosexual and heterosexual samples. The Rorschach, Bell Object Relations Inventory (BORI), and Adult Attachment Scale (AAS) were administered to 120 adult male volunteers (60 homosexual and 60 heterosexual) classified into three relationship types as described by Bell and Weinberg (1978): Close-Coupled, Single Functional, and Single Dysfunctional. Results of MANOVA revealed that Close-Coupled and Single Functional groups scored significantly lower on the BORI subscales of Alienation, Insecure Attachment, Egocentricity, and Social Incompetence, and on the AAS Anxiety scale, as well as scoring higher on the AAS Close and Depend scales, than did the Single Dysfunctional group. No significant differences were obtained between homosexual and heterosexual groups on these scales, except for Anxiety. The homosexual mean score on this scale was significantly higher than the heterosexual mean. Human content on the Rorschach was scored according to the Concept of the Object Scoring System (COSS; Blatt et al, 1976) for: differentiation, articulation, and integration. No significant differences were revealed for COSS scores across commitment/satisfaction levels, or sexual orientation groups, with the exception of homosexual subjects obtaining a significantly higher mean score on Articulation of inaccurately perceived human responses, a finding that could be due to chance given the number of tests. No significant interactions between sexual orientation and commitment/satisfaction were found on any of the dependent measures, indicating that these factors were independent. COSS scores were uncorrelated with BORI subscales, suggesting either COSS scores were insufficiently sensitive to actual levels of object relations pathology found in this sample, or that BORI scores overestimated psychopathology. The findings were interpreted as generally supporting the predictions that object relations and attachment pathology would vary with relationship commitment/satisfaction, and not related to homosexuality per se