Musical expectations—and their subsequent denial or confirmation—have been identified as a powerful mechanism of musical experience. In music cognition research, such expectations are often treated deterministically: given a listener’s global listening history and a context-providing stimulus, a listener will make a particular prediction. Yet, listeners may also be able to intentionally modulate their expectations based on how they are prompted to interpret musical stimuli, a phenomenon discussed in music theory as “hearing as.”

I present two neuroscientific experiments which investigate such interpretation effects. The first shows that knowledge of the syntactic structure of a stimulus modulates neural signals for surprisal: in other words, just how surprising an improbable event is depends on how a listener is integrating it into the local context. The second experiment directly instructs participants to adopt different “listening stances” while listening to perceptually bi-stable chords ([0 4 7 10] can be a V⁷or a Ger⁺⁶ chord). Independent of global statistics and local context, such listening stances can modulate prediction errors, such that even probable continuations (like V⁷ --> I) can evidently be made to evoke a prediction error signal (e.g., if the participant listens to the chord preceding I as a Ger⁺⁶ chord.)

These findings add a layer of complication to neuroscientific studies of musical expectation, prompting greater consideration of how listeners are actually engaging with experimental stimuli and tasks, and highlighting the need to consider how yet other forms of interpretation may modulate predictions and its neuroscientific theorization.

Andrew Goldman is Assistant Professor in Music Theory and Cognitive Science at Indiana University, where he also directs IU's Music and Mind Lab. He received his PhD in 2015 at the Centre for Music and Science at the University of Cambridge. He was in the inaugural cohort of Presidential Scholars in Society and Neuroscience at Columbia University, and also held a postdoctoral position with the Music, Cognition, and the Brain initiative at Western University before joining the faculty at Indiana University. His research considers how we can (and cannot) use scientific methods to learn about musical perception and cognition in theory, and he also designs and conducts behavioral and neuroscientific experiments on music perception and cognition. Goldman's research has focused on improvisation in music and dance, and more recent work has addressed the perception of musical form, melodic contour, metrical perception, embodiment in music, harmonic expectation, and a critique of music neuroscience. His work has been published in both music and psychology journals, and has been presented at national and international conferences. Goldman is also a pianist and composer. His original musical, "Science! The Musical" has been produced four times, and provides an alternate platform to explore the worlds of music and science.