Does order matter? Harmonic priming effects for scrambled tonal chord sequences.

Abstract

This study examines whether scrambling the order of events in a tonal chord sequence inhibits the speed and accuracy of processing in two behavioral harmonic priming experiments. Sixteen 9-chord sequences were adapted from Bach’s chorales that either remained unchanged (thereby reflecting high temporal coherence) or were scrambled to produce increasingly incoherent sequences (i.e., medium or low). To produce the scrambled conditions, a finite-context (or n-gram) model trained on a corpus of chord annotations and then identified the scrambled versions (scrambling chords 2–8) that produced high estimates of model surprisal. In both experiments, 60 participants (30 musicians) indicated as quickly as possible whether the target chord was in or out of tune, where out-of-tune trials were either fixed at a tuning level of 40 cents sharp relative to the preceding context (Experiment 1), or at a tuning level representing the intonation discrimination threshold of each participant, which was estimated using an adaptive staircasing procedure before the main session began (Experiment 2). Correct response times and sensitivity measures replicated the high-to-low staircase found in the model estimates, suggesting harmonic priming effects reflect the order of chords in a sequence. Implications for topological and temporal models of tonal-harmonic structure are discussed.

When in Rome: A Meta-corpus of Functional Harmony

“…Sears et al (2023)'s selection of 100 Bach Chorales is encoded in a kind of hybrid, with the DCML regex syntax inside a kern spine.…”