Search

ABSTRACTAs a concept, romantic intimacy may be examined through Schaefer and Olson's (1981) measure, the Personal Assessment of Intimacy in Relationships (PAIR), which identified five dimensions of intimacy: emotional, intellectual, recreational, sexual, and social intimacy. Moore, McCabe, and Stockdale's (1998) attempt to replicate this five-factor model was not successful, and they instead proposed a three-factor model of intimacy: engagement, communication, and shared friendships. The objective of the present study was to examine the psychometric properties of the PAIR with individuals involved in same-sex couple relationships using these two models. Three hundred fifty participants completed the PAIR, the Experiences in Close Relationships, and the brief 4-item version of the Dyadic Adjustment Scale. Confirmatory factor analyses were not able to replicate both the original five-factor structure and the proposed three-factor structure of the PAIR. Rather, results supported the factor structure, reliability, and concurrent validity of our revised version of Moore et al.'s (1998) three-factor model of intimacy, which retained the three dimensions after various modifications (i.e., addition of covariance terms, deletion of an item, and transference of items) were made. Links were found between these three dimensions of intimacy and insecure romantic attachment, as well as relationship happiness.

We recently reported that, among bulimic women, previously abused carriers of the 5HTTLPR S allele showed special propensities towards novelty seeking (implying recklessness or impulsivity) and interpersonal insecurity. We subsequently re‐analyzed our data, to examine the bearing of the 5HTTLPR polymorphism and prior sexual or physical maltreatment upon validated, higher‐order personality‐traits. Ninety women with bulimic syndromes were genotyped for 5HTTLPR “short” (S) and “long” (LG and LA) alleles, and then assessed for eating symptoms, history of sexual or physical abuse, and the higher‐order personality traits Emotional Dysregulation, Dissocial Behavior, Inhibition, and Compulsivity. With a classification based on a biallelic model of 5HTTLPR (i.e., presence or absence of at least one S‐allele copy), multiple regression indicated a significant proportion of variance in Dissocial Behavior to be explained by an abuse × genotype interaction—greater psychopathology occurring in abused S‐allele carriers. A parallel analysis applying a triallelic model of 5HTTLPR (i.e., presence or absence of at least one copy of presumably low‐function S or LG alleles) produced a similar pattern, but no statistically significant effect. The finding that bulimic 5HTTLPR S‐allele carriers who are previously abused display elevations on Dissocial Behavior corroborates previous observations concerning phenomenological correlates of traumatic stress in 5HTTLPR S allele carriers. © 2007 Wiley‐Liss, Inc.