Tonal expectations and pitch discrimination

Musical expertise is not limited to the explicit knowledge that results from musical training, it also includes an implicit knowledge of the tonal system that is acquired through mere exposure to music (Francès, 1958; Krumhanl, 1990). This implicit knowledge accounts for a number of musical abilities that do not depend on formal musical training (Bigand & Poulin-Charronnat, 2006). The present study investigates the influence of this implicit knowledge on pitch discrimination as revealed by the ERPs elicited. A first step in this direction has been made by Brattico, Näätänen, and Tervaniemi (2002), who compared musicians and non-musicians listeners’ ERPs to infrequent note changes within a tonal context (the five first tones of the A major scale) or within a non-tonal context (five tones that did not reproduce any tonal scale). The note change modified the melodic contour and, for tonal contexts, shifted the A-major mode to an A-minor mode. Participants had to read a book and ignore the stimuli. Pitch changes elicited larger MMNs in the tonal context condition than in the non-tonal context condition, in both musicians and non-musicians. This result reflects the influence of musical acculturation: a tonal context, which observes the regularities of Western music, helps processing pitch changes, even pre-attentively. However, listeners’ implicit knowledge is subtler than just differentiating tonal and non-tonal contexts, and small differences in tonal contexts have been shown to influence pitch discrimination. In a behavioral study (Marmel, Tillmann, & Dowling, 2008), we showed that a single note change in melodic contexts that changed the tonal function of the last tone of the melodies influenced processing of this last tone’s pitch. In this study, pairs of melodic contexts were constructed so that the two melodies of a pair differed only by one (possibly repeated) tone in the beginning of the melodies. This tone change caused the two melodies to be in different tonalities and thus the last tone to have a different tonal function in the two melodies. In one melody of a pair, the last tone was the tonic (most tonally expected tone), and in the other melody it was the subdominant (less tonally expected). This single tone change allowed us to keep identical the sensory components of expectations, like melodic contour or tone repetition, between the two melodies of a pair. Thus, this design allowed us to focus on cognitive expectations linked to listeners’ implicit tonal knowledge. In one experiment, the melodies were presented with their last tone being repeated identically or slightly lowered in pitch, and participants had to judge whether the last tone was repeated identically or not. Results showed better pitch discrimination when the last tones were more tonally expected (i.e., tonic) than when they were less tonally expected (i.e., subdominant) (Marmel, Tillmann, & Dowling, 2008, Exp.3). The present study will use the same melodies and task in an ERP experiment. The repetition of the last tone should allow us to observe different ERPs on penultimate and final tones: Penultimate tones should be associated with components elicited by the tonal deviation processing whereas final tones should be rather associated with components elicited by pitch deviation processing and task-related decisional processes.