ABSTRACTBackground:Impairments in facial emotion recognition (FER) have been detected in patients with Parkinson disease (PD). Presently, we aim at assessing differences in emotion recognition performance in PD patient groups with and without mild forms of cognitive impairment (MCI) compared to healthy controls.Methods:Performance on a concise emotion recognition test battery (VERT-K) of three groups of 97 PD patients was compared with an age-equivalent sample of 168 healthy controls. Patients were categorized into groups according to two well-established classifications of MCI according to Petersen's (cognitively intact vs. amnestic MCI, aMCI, vs. non-amnestic MCI, non-aMCI) and Litvan's (cognitively intact vs. single-domain MCI, sMCI, vs. multi-domain MCI, mMCI) criteria. Patients and controls underwent individual assessments using a comprehensive neuropsychological test battery examining attention, executive functioning, language, and memory (Neuropsychological Test Battery Vienna, NTBV), the Beck Depression Inventory, and a measure of premorbid IQ (WST).Results:Cognitively intact PD patients and patients with MCI in PD (PD-MCI) showed significantly worse emotion recognition performance when compared to healthy controls. Between-groups effect sizes were substantial, showing non-trivial effects in all comparisons (Cohen's ds from 0.31 to 1.22). Moreover, emotion recognition performance was higher in women, positively associated with premorbid IQ and negatively associated with age. Depressive symptoms were not related to FER.Conclusions:The present investigation yields further evidence for impaired FER in PD. Interestingly, our data suggest FER deficits even in cognitively intact PD patients indicating FER dysfunction prior to the development of overt cognitive dysfunction. Age showed a negative association whereas IQ showed a positive association with FER.
Moral processing deficit in behavioral variant frontotemporal dementia is associated with facial emotion recognition and brain changes in default mode and salience network areas
