Inter-subject Correlation While Listening to Minimalist Music: A Study of Electrophysiological and Behavioral Responses to Steve Reich's Piano Phase

Abstract

Musical minimalism utilizes the temporal manipulation of restricted collections of rhythmic, melodic, and/or harmonic materials. One example, Steve Reich's Piano Phase, offers listeners readily audible formal structure with unpredictable events at the local level. For example, pattern recurrences may generate strong expectations which are violated by small temporal and pitch deviations. A hyper-detailed listening strategy prompted by these minute deviations stands in contrast to the type of listening engagement typically cultivated around functional tonal Western music. Recent research has suggested that the inter-subject correlation (ISC) of electroencephalographic (EEG) responses to natural audio-visual stimuli objectively indexes a state of "engagement," demonstrating the potential of this approach for analyzing music listening. But can ISCs capture engagement with minimalist music, which features less obvious expectation formation and has historically received a wide range of reactions? To approach this question, we collected EEG and continuous behavioral (CB) data while 30 adults listened to an excerpt from Steve Reich's Piano Phase, as well as three controlled manipulations and a popular-music remix of the work. Our analyses reveal that EEG and CB ISC are highest for the remix stimulus and lowest for our most repetitive manipulation, no statistical differences in overall EEG ISC between our most musically meaningful manipulations and Reich's original piece, and evidence that compositional features drove engagement in time-resolved ISC analyses. We also found that aesthetic evaluations corresponded well with overall EEG ISC. Finally we highlight co-occurrences between stimulus events and time-resolved EEG and CB ISC. We offer the CB paradigm as a useful analysis measure and note the value of minimalist compositions as a limit case for the neuroscientific study of music listening. Overall, our participants' neural, continuous behavioral, and question responses showed strong similarities that may help refine our understanding of the type of engagement indexed by ISC for musical stimuli.

Keywords: EEG; continuous behavioral measure; engagement; inter-subject correlation (ISC); minimalism (music); music cognition.

Figure 1

The opening modules from Steve Reich's Piano Phase. Lines under the staff indicate sections: blue lines are in-phase sections and red lines are phasing sections.

Figure 2

The waveforms for each of the stimuli in the experiment. (A) Original, with phasing sections colored gray and the progression of events represented by the gradual change of color from white to blue. (B) Abrupt Change, white lines denoting sudden shift from one in-phase section to the next and background color showing approximate location of in-phase material in the Original condition. (C) Segment Shuffle, random re-ordering of 5-s units shown using original color in Original. (D) Remix [Winn's Piano Phase (D*Note's Phased & Konfused Mix)], gradual progression of events represented with color change from gray to yellow and key musical events beginning with white lines. (E) Tremolo, appearing as an unchanging block when zoomed out, but in the lower plot, zoomed in to show the reiterated pitch material.

Figure 3

Analysis pipeline for experiment data. Participants heard each of the five stimuli twice, once in each block. During Block 1 we recorded EEG, and during Block 2 participants completed the continuous behavioral (CB) task. Participants answered questions about each stimulus after hearing it. For the EEG data we computed spatial components maximizing temporal correlation and projected electrode-by-time response matrices to component-by-time vectors. For vectorized EEG as well as CB vectors, we then computed inter-subject correlation (ISC) of the vectors on a per-stimulus basis, across time and in a time-resolved fashion. We additionally computed the time-resolved mean values between participants. We aggregated and analyzed ratings.

Figure 4

Behavioral ratings for all questions in the experiment (responses were ordinal and are slightly jittered for visualization only). Ratings for “pleasant,” “musical,” “well ordered,” and “interesting” come from Block 1 and ratings for “engaging” come from Block 2. For pleasant, musical, interesting, and engaging, responses for Remix were significantly higher than for the other conditions. For these same questions, responses were also significantly lower for Tremolo compared to all other conditions. For ratings of well ordered, we saw a similar pattern except that Abrupt Change was significantly higher than Segment Shuffle. Asterisks denote significance of p < 0.05. Please refer to the online version of the paper for the full-color figure.

Figure 5

EEG components, coefficients, and aggregate ISC. (A) Spatial filter weights are visualized on a scalp model using forward-model projections. Maximally reliable components (RC1) exhibit consistent auditory topographies for all stimulus conditions except Tremolo. (B) Spatial filter eigenvalues serve as component coefficients. Significant coefficients are marked with red asterisks and significance thresholds; gray areas denote the 95th percentile of the null distribution. RC1 is statistically significant for all conditions except Tremolo. (C) ISC was computed over the entire duration of each stimulus. Remix elicited significantly higher ISC than all the other conditions, and Tremolo elicited significantly lower ISC than all other conditions. Individual participants' EEG ISC values are denoted with dots. Gray areas denote the 95th percentile of the null distribution. Asterisks denote significance of p < 0.05.

Figure 6

ISC of continuous behavioral (CB) reports of engagement for each condition with individual participant data and standard error of the mean plotted. Shaded gray regions denote the 95th percentile of the null distribution. Remix elicited significantly higher ISC than all the other conditions and Tremolo elicited significantly lower ISC than all the other conditions. Segment Shuffle also differs significantly from all other conditions. Asterisks denote significance of p < 0.05.

Figure 7

Time-resolved EEG ISC, CB ISC, and CB means for each condition. The top of each shaded gray region represents the 95th percentile of the corresponding null distribution. (A) Original: Dotted lines mark the start of phasing sections, solid lines mark the start of in-phase sections. (B) Abrupt Change: Solid lines mark the start of each new in-phase section. (C) Segment Shuffle: Light gray lines mark the start of each new segment. (D) Remix: Dashed lines mark musical events expected to be significant to listeners. (E) Tremolo.