The authors of “Altered Function and Connectivity of the Medial Frontal Cortex in Pediatric Obsessive-Compulsive Disorder”, published in Biological Psychiatry (2010;68:1039–1047), have discovered an inaccuracy regarding how the fMRI analysis was conducted (see Data Analysis section, page 1040).
        Specifically, Fitzgerald et al. stated that only correct trials were used in the interference contrast (incongruent vs. congruent conditions), but in fact, both correct and incorrect response types for each condition were included. The obsessive-compulsive disorder (OCD) and healthy control subjects did not differ with respect to accuracy, so the number of error trials included in the interference analysis did not differ between groups.
        Upon discovery of this error, the authors performed a re-analysis of incongruent vs. congruent conditions with correct trials only, as indicated by the original statement. The authors also now provide a summary of these new results:
        • 1
          The originally reported dorsal anterior cingulate hyperactivation for OCD compared to healthy controls appears only at lower thresholding.
        • 2
          As originally reported, OCD patients show hyperactivation of the ventral medial frontal cortex; however, this result localizes to rostral anterior cingulate cortex, rather than medial frontal pole.
        • 3
          The new interference analysis (correct trials only) shows one new finding: bilateral dorsolateral prefrontal cortex hyperactivation in OCD. The authors observed the same finding in the original analysis at lower thresholding.

        Linked Article