Verbatim repetition and musical engagement.

The topics of musical similarity and variation have received considerable attention both from a psychological perspective (Deliege, 2007) and from a musicological perspective (Sisman, 1993; Zbikowski, 2002). Less well studied has been its correlate, exact repetition. A recent book (Margulis, 2013a) attempts to remedy that oversight, synthesizing work from a number of fields to outline a theory about the enduring appeal of musical repetition. Recent empirical work has suggested that repetition can shift listeners' attention to higher-level attributes of the musical stimulus (Margulis, 2012) and elicit more enjoyment from listeners and higher ratings of intentional artistry (Margulis, 2013b).An interesting borderline case between the poles of repetition and variation is performer nuance. Even when the notation is identical among different iterations of a principal theme, performers often change expressive inflections, altering dynamics and microtiming on each rendition. For example, the famous theme in Chopin's Polonaise in A major, Op. 40/1 recurs several times across the course of the piece. Pianists make many nuanced changes from rendition to rendition. For example, a pianist can withhold volume during the penultimate rendition to make the unleashing of maximal volume during the final rendition even more impressive.A skeptic might inquire whether these differences are intentional; perhaps performers simply lack the cognitive or motor control to execute sequences with a high degree of expressive fidelity. Gabrielsson (1987) examined pianists' expressive timing as they performed the first movement of Mozart's Piano Sonata in A, K. 331. Figure 1 captures the percentage deviation from notated durational values in performances by two pianists of the theme on its first statement (solid line) and then on its subsequent repeat (dotted line). The dotted line cleaves closely to the solid line for both pianists; their second performance of the theme was quite expressively faithful to their first. Shaffer (1984) also provides evidence that performers are capable of reproducing expressive elements. That study measured expressive timing characteristics during two performances of a Chopin Etude recorded two years apart by the same pianist. The second performance replicated the first in remarkable detail.Performers are able to replicate expressive inflections from one performance to another, so it seems likely that the deviations they make are expressively intentional. The question of what motivates these expressive choices is a fascinating topic of ongoing research (cf. Juslin & Timmers, 2010). An associated issue is the effect of these replications and deviations on listeners. Are they able explicitly or implicitly to register changes in performer nuance across renditions? How much deviation is required before a listener registers it? Do they prefer performances where repetitions are mostly exact, or performances where subtle variations have been introduced? What are the consequences of exact versus varied repetition for other aspects of musical listening and attention?Repetition, unsurprisingly, tends to enhance memory for melodies. Deutsch (1979) tested people's memory for a short atonal melody after it had been presented in one of several ways: a single time, repeated such that the second statement was an exact repetition of the first, or repeated such that the second statement involved an octave transposition. Performance on the memory task was best when the participants had heard the melody repeated exactly, but even a transposed repetition improved performance better than did a single presentation. Gardiner, Kaminska, Dixon, and Java (1996) showed that repetition increased both remember and know responses for music in an unfamiliar style, and remember but not know responses for music in a familiar style.Studies have shown that memory for the verbatim aspects of speech-the exact words and sounds used to convey the thoughts- decays rapidly and is replaced by memory for the gist, the basic content of the utterance (Gernsbacher, 1985; Reyna & Brainerd, 1995). …