The low correlations reported between self-report measures and the Rorschach raise questions about the validity of both kinds of instruments. Meyer (1996) suggested that these low correlations are an artefact, due to the failure to control response style. Correlations would be high and positive when subjects have the same response style on both methods, and high and negative when they have divergent response styles. But response style is assessed according to criteria strongly connected with distress, and the enhancement of correlations may be tautologically limited to distress scales. The objective of this research is to verify whether the response style hypothesis applies to Openness to Experience, a dimension unrelated with distress. Correlations were computed between on the one hand Openness and Neuroticism, and on the second one several selected Rorschach variables. Analyses were conducted on the whole sample (n = 96) and on separate subgroups of patients with convergent (n = 29) or divergent (n = 22) test-taking attitude. The findings establish Openness is related to Rorschach low L, high blends, morbid responses, and combinatory special scores. They also establish that response style moderates correlations between Neuroticism and the Rorschach, but not between Openness and the Rorschach. These results seem to confirm the tautological nature of the present formulation of the response style hypothesis, but further refinement in the analysis of test-taking attitude could help reconsider things.