Flipbook

4 SAFETY BEHAVIOR INCREASES OBSESSION-RELATED COGNITIONS 109 Obsessive beliefs questionnaire-44 (OBQ-44) responsibility and threat estimation (RT) scale There were no main effects of Time, F (1,86) = 0.70, p = .41, and Condition, F (2,86) = 0.48, p = .62, nor was the Time x Condition interaction significant, F (2,86) = 0.83, p = .44. Thus, in contrast with the findings on the CCS and with our expectations, the experimental group did not show a pre- to post-test increase in obsessive beliefs about inflated responsibility and exaggerated threat perceptions compared to the monitor and control group. Beck anxiety inventory (BAI) There was a main effect of Condition, F (2,86) = 3.55, p = .03, η p ² = .08, and an overall decrease in BAI scores between pre- and post-test, F (1,86) = 4.79, p = .03, η p ² = .05. There was a trend for this to differ between conditions, F (2,86) = 2.85, p = .06, η p ² = .06. This pattern of differences between conditions was further investigated with exploratory post-hoc tests. These indicated a decrease in the monitor group, t (28) = 2.78, p = .01, d = 0.47, but no change in the experimental, t (29) = 0.35, p = .73, and control group, t (29) = 1.68, p = .10, see Table 1. The change in BAI scores differed between the experimental and monitor group, t (57) = 2.14, p = .04, d = 0.56, but not between the experimental and control group, t (48.58) = 1.19, p = .24, and the monitor and the control group, t (57) = 1.38, p = .18. This was in line with the expectation that the effect of the checking behavior manipulation would be OCD-specific, and thus would not increase general anxiety. DISCUSSION Increasing daily checking behavior of everyday objects for one week increased overall checking-related cognitions measured with the CCS, which assessed cognitions about the importance of checking and the likelihood and severity of threat (i.e., threat overestimation). Moderate to strong correlations between these constructs