The facial emotion recognition deficit in Parkinson's disease: Implications of a visual scanning strategy

Objective: We explored the relationship between a visual scanning strategy and a facial emotion recognition deficit in Parkinson's disease (PD). Method: Thirty nondemented PD patients (balanced for symptom side at onset) and 20 age, education and gender-matched healthy controls (HC) were enrolled. The PD group underwent a comprehensive neuropsychological battery also exploring the executive functions. In both groups, eye movements were recorded while subjects categorized facial emotion from Ekman's 60-faces test. We were particularly interested in the location of fixations on facial pictures (top vs. bottom) and in emotional valence (positive vs. negative). We also compared performance of the two groups on a verbal emotion attribution task. Results: Compared to HC, PD patients performed worse on visual recognition of negative emotions such as anger, fear, and sadness (where the upper part of the face is more informative than the lower part); the two groups did not differ on the verbal emotion attribution task. HC modified their visual scanning strategy (both number and overall time duration of fixations) according to the valence of the emotion; by contrast, PD showed the same pattern regardless of the valence. In the PD group, accuracy in the visual recognition of negative emotions and fixation pattern correlated with performance on tasks exploring executive functions; however, no associations were observed with severity of motor state. Conclusions: Our results suggest that visual scanning strategy contributes significantly to the facial emotion recognition deficit of PD patients, especially at a "high level" related to cognitive control of eye movements.