Abstract
Body dysmorphic disorder (BDD) is characterized by distressing or impairing preoccupation with imagined or slight defects in patients' own appearance, especially faces. Previous studies have suggested that distortion of holistic visual processing is involved in BDD. This raises the possibility that training particularly on low spatial-frequency components of images would at least partially reduce the symptoms of BDD. Here, we tested whether visual processing of low spatial-frequency with BDD patients is different from normal controls either before or after training on low spatial-frequency components of images of faces. Individuals with BDD (n=9) as well as controls (n=10) were trained for 6 days on a two-interval forced choice task. Participants underwent fMRI sessions before and after training while performing the two-interval forced choice task. Performance on faces and houses in low (< 10 cycle per stimulus) and high (> 15 cycle per stimulus) spatial-frequency bands was assessed during scans. We found no qualitative difference in performance between the patients and controls either before or after training: training made both groups selectively increase their sensitivity to low spatial-frequency faces (low spatial-frequency face: p< 0.01, other conditions: p>0.1). However, before training, there was a significant difference in brain processing: In normal controls, fMRI responses in FFA were stronger in the right than left hemispheres. The degree of the right dominance was smaller in BDD patients than controls. After training, however, BDD patients showed increased right dominance while normal controls showed reduced right dominance, so that the difference in the degree of right dominance between the two groups was washed out. These patterns were not observed in early visual areas or in the parahippocampal place area. We conclude that the neural mechanisms of visual processing and plasticity are different between BDD patients and normal controls, although no significant behavioral difference is shown.
Meeting abstract presented at VSS 2017