The present study investigated whether MI is a mechanism for change in the treatment of obsessive–compulsive disorder (OCD). The Magical Ideation scale (MI), the Obsessive–Compulsive Inventory — Short Version (OCI-SV) and the Padua Inventory were completed by 34 obsessive–compulsive patients pre- and post cognitive–behavioural treatment. Treatment did not target magical styles of thinking. Significant improvements on all three measures of obsessive–compulsive symptoms were demonstrated by t tests over the course of treatment. Improvement in magical thinking was also shown to be significant in t test results. In support of the hypothesis, correlations between MI improvement and improvement on the obsessive–compulsive symptom scales were significant (at a level of .05) suggesting that there is an association between improvement in magical thinking and improvement in obsessive–compulsive symptoms. Notably, a significant negative correlation was obtained between prescores on MI and change scores on the OCD measures. This suggests that high levels of MI are associated with high levels of treatment intractability. High MI appears to be a poor prognostic factor in OCD.

Data collected from clinical populations indicate that magical ideation (MI) may play a causal or a mediating role in the expression of obsessive compulsive symptoms. If this is the case then when targeted in treatment, symptoms of obsessive compulsive disorder (OCD) should be altered. Two individuals diagnosed with OCD received a trial treatment targeting magical thinking. The intervention consisted of a series of procedures designed to undermine superstitious/MI without targeting obsessions or compulsions. The procedures involved critical analysis of the following material: (1) a free astrology offer; (2) a horoscope prediction exercise; (3) a description of four different cultural explanations of the origin of fire; (4) an instructive guide for Tarot card readers; (5) a report of a UFO sighting; (6) a video-clip describing a cult festival; (7) a description of a ‘hoax’ channeler and (8) a superstition exercise. Measures of obsessive compulsive symptoms, superstition, MI and thought–action fusion were administered pre-treatment, post-treatment and at 3 months’ follow-up. According to the twofold criterion of Jacobson et al. (Behaviour Therapy 1984, 15, 336–352), following treatment the patients were identified as being recovered on measures of magical and superstitious thinking and on the Padua Inventory.