Pre-attentive processing of neutral and emotional sounds in congenital amusia

Abstract

Congenital amusia is a life-long deficit of musical processing. This deficit can extend to the processing of language and in particular, emotional prosody. In a previous behavioral study, we revealed that while amusic individuals had difficulties in explicitly recognizing emotions for short vowels, they rated the emotional intensity of these same vowels as did their matched control participants. This finding led to the hypothesis that congenital amusics might be impaired for explicit emotional prosody recognition, but not for its implicit processing. With the aim to investigate amusics automatic processing of prosody, the present study measured electroencephalography (EEG) when participants listened passively to vowels presented within an oddball paradigm. Emotionally neutral vowel served as the standard and either emotional (anger and sadness) or neutral vowels as deviants. Evoked potentials were compared between participants with congenital amusia and control participants matched in age, education, and musical training. The MMN was rather preserved for all deviants in amusia, whereas an earlier negative component was found decreased in amplitude in amusics compared to controls for the neutral and sadness deviants. For the most salient deviant (anger), the P3a was decreased in amplitude for amusics compared to controls. These results showed some preserved automatic detection of emotional deviance in amusia despite an early deficit to process subtle acoustic changes. In addition, the automatic attentional shift in response to salient deviants at later processing stages was reduced in amusics in comparison to the controls. In the three ERPs related to the deviance, between-group differences were larger over bilateral prefrontal areas, previously shown to display functional impairments in congenital amusia. Our present study thus provides further understanding of the dichotomy between implicit and explicit processing in congenital amusia, in particular for vocal stimuli with emotional content.