Facial, prosodic, and lexical emotional perception in unilateral brain-damaged patients

This study examined emotional perception in unilateral brain-damaged subjects. The uniqueness of this project is that emotional perception was evaluated in three communication channels: facial, prosodic, and lexical. Subjects were 11 right brain-damaged (RBD), 10 left brain-damaged (LBD), and 15 normal control (NC) adults. The RBDs and LBDs were included for study if they suffered brain damage as a result of a unilateral CVA. All subjects were native English speakers or were fluent in English by the age of seven. Subjects did not have any history of mental retardation, psychotropic drug treatment, psychiatric disorder, or substance abuse. Subject groups were comparable with regard to age, education, and occupational level. All subjects received a battery of screening, control, and experimental measures. The screening measures were used to ensure a minimum level of cognitive and sensory functioning. Within each channel, nonemotional tasks were employed to control for factors that could influence performance on the experimental emotional tasks. For the facial channel, neutral face recognition was measured. For the prosodic channel, the ability to process intonation contours was assessed. Within the lexical channel, a nonemotional word identification task, a nonemotional word discrimination task, and a nonemotional sentence identification task were used. For the experimental tasks, subjects identified the emotion represented by facial, prosodic, and lexical stimuli. Both identification and discrimination tasks were employed, and positive and negative emotions were used. For the identification tasks, significant group differences occurred, such that RBDs were impaired relative to LBDs and NCs across the three channels, regardless of valence. No subject-group differences were found for the discrimination tasks. Further, when performance on lexical tasks was examined, RBDs were significantly more impaired on emotional than on nonemotional tasks. When correlation coefficients were computed among channels, the magnitude of the correlations was greater for identification than for discrimination tasks, suggesting the possibility of a general emotional processor for identification. Overall, the results of the present study support the right hemisphere hypothesis for emotion. Future studies should include a larger number of subjects so that other factors (e.g., intrahemispheric lesion location) could be examined.