Implicit self- and other-associations in obsessive-compulsive personality disorder traits

Abstract

This study investigated automatic associations with the self and with others in the context of obsessive-compulsive personality disorder (OCPD). Participants scoring high (n=20) or low (n=20) on DSM-IV OCPD traits performed three irrelevant-feature tasks in which language of the words (Dutch vs. English) was the relevant stimulus feature and word content (OCPD self-view, OCPD other-view, high self-esteem and low self-esteem) the irrelevant feature. Results showed that the high and low OCPD group differed with respect to both explicit and implicit OCPD cognitions. Typically, high OCPD participants showed better performance on OCPD-congruent trials than on OCPD-incongruent trials, whereas low OCPD participants displayed the opposite pattern. This was evident from a semantic Simon effect and from a semantic priming effect. Correlations between direct and indirect measures of OCPD beliefs and of self-esteem were low. Moreover, the indirect OCPD measures contributed uniquely and independently of explicit beliefs to the prediction of OCPD.

Acknowledgements

Anoek Weertman is now at Viersprong Institute for Studies on Personality Disorders, Halsteren, The Netherlands. Peter de Jong is now at Department of Developmental and Clinical Psychology, Section Experimental Psychopathology, University of Groningen, The Netherlands. Mike Rinck is now at the Behavioural Research Institute at Radboud University Nijmegen, The Netherlands.

We thank Bert Hoekzema for his useful suggestions and his technical assistance during the preparation of the experiment and data processing. We thank Astrid Dello for her assistance during data acquisition and processing, and the reviewers for helpful comments on an earlier version of this article.

Notes

¹In fact, there is yet another compatibility effect, namely compatibility of prime language and target language: Since all primes were presented in Dutch, responses to Dutch targets should be faster than responses to English targets. We did not analyse this effect, however, because it is not of theoretical interest, it is confounded with language proficiency, and because we only analysed responses to Dutch targets.

²Indeed, additional analyses of the English stimulus words did not reveal systematic effects. There were only significant main effects of Word Type (Task 1: F=23.5, p≤.001; Task 3: F=55.8, p≤.001) or interaction effects of Word Type and Prime (Task 2: F=21.5, p≤.001).