Perceptual and neural processing of consonance and dissonance in musicians and non-musicians: an ERP study

The concept of consonance and dissonance, which are considered to evoke pleasant and unpleasant emotions is one of the most important fundamental concepts in the theory of music. Some argued that perceiving the consonance and dissonance is innate (Trainor and Trehub, 1993) and others argued that expertise in music can influence the perception of consonance and dissonance (Kameoka and Kuriyagawa, 1969). Previous studies attempted to understand neural processing of sensory consonance and dissonance and to see if musical expertise affects the neural processing by measuring event-related potentials (ERPs) of musicians and non-musicians presented with consonant and dissonant chords out of musical context (Itoh et al. 2003; Minati et al. 2009). Although these studies demonstrated different ERPs to the consonant and dissonant chords and differences in ERPs between musicians and non-musicians, the results were contradictory and needed to be verified. The present study proposes that interstimulus intervals used in the previous studies were inadequate to separate one chord from the other in order to avoid local context effect, which occurs by the sensory traces of the stimuli that precede it. The aims of this study were (1) to investigate the influence of musical expertise in the perceiving consonance and dissonance; (2) to verify if the interstimulus interval is the cause of the previous studies’ contradictory results and if so, (3) to investigate the influence of local context effect on the neural process of consonance and dissonance. Four different experimental blocks containing consonant and dissonant chords were presented to 24 participants (12 musicians and 12 non-musicians) and subjective ratings and ERPs were obtained. The first and fourth blocks (named ‘mixed blocks’) contained 33 consonant and 33 dissonant chord stimuli presented in random orders. The second and the third blocks (named ‘pure blocks’) contained 66 consonant chords or 66 dissonant chords, respectively. Interstimulus interval was 3.5 seconds - a little longer than that of the previous studies but shorter than the duration of human echoic memory trace, which is 4 seconds (Watkins and Watkins, 1980). Both musicians and non-musicians rated dissonant chords as unpleasant, and non-musicians rated consonant chords as pleasant whereas musicians rated them as neutral. The N1 ERP derived from electroencephalography (EEG) was larger for consonant chords than dissonant chords in both groups. This was observed in frontocentral brain regions for non-musicians and in centroparietal brain region for musicians. P2 amplitude was larger for consonant chords than dissonant chords in pure block conditions for musicians only. The consonant and dissonant stimuli effect in musicians was found in the pure block but not in the mixed block condition, indicating presence of the local context effect. The N2 component was larger for dissonant chords than consonant chords in the pure blocks for both musicians and non-musicians but the N2 modulation was observed only in frontal regions of musicians’ brains and distributed throughout frontal and central regions of non-musicians’ brains. In this novel behavioural and electrophysiological study, there was evidence of similarities and differences in perceptual and neural processing of consonance and dissonance between musicians and non-musicians. Perception of consonance and dissonance was universal but musicians’ enhanced listening ability influenced the perception of consonance and dissonance. Neural processing of consonance and dissonance was universal to some extent but the enhanced listening ability due to musical expertise has contributed to the changes in the regional brain function. Furthermore, the result provides evidence that the inadequate interstimulus intervals produced the local context effect