Musical structure processing after repeated listening: Schematic expectations resist veridical expectations

The present study investigates whether expectations based on listeners’ schematic knowledge of music (related to tonal functions and reflected in harmonic priming) may be modulated by veridical expectancies (knowing what to come, here tested with repetition priming). It is well established that target chord processing is facilitated when the target acts as the most referential chord of the Western musical system (i.e., the tonic chord) in the prime context. Our study attempted to modulate this harmonic priming effect by presenting half of participants with numerous sequences ending on a moderately referential chord (i.e., the subdominant chord) before the experimental test. The second half of participants was presented with sequences ending on the most referential tonic chord. This repeated processing was supposed to modulate the strength of the harmonic priming effect, with lesser priming in the former than in the latter group. These sequences were either different idioms than those in the exposition phase, but with the same musical structure (Experiment 1) or identical repetitions (Experiment 2). Harmonic priming effects were observed in both experiments, and were only weakly affected by repetition priming in Experiment 2. This outcome underlines the strength of schematic expectations, which resist veridical expectations, and it provides some empirical ground for the role of expectations in musical expressivity despite listeners’ knowing about what will come next.