Scale-sensitivity: A cognitive resource basic to music perception

Abstract

A tone-scramble is a rapid, randomly ordered sequence of pure tones. Chubb, Dickson, Dean, Fagan, Mann, Wright, Guan, Silva, Gregersen, and Kowalski [(2013). J. Acoust. Soc. Am. 134(4), 3067–3078] showed that a task requiring listeners to classify major vs minor tone-scrambles yielded a strikingly bimodal distribution. The current study sought to clarify the nature of the skill required in this task. In each of the “semitone” tasks, all tone-scrambles contained eight each of the notes G5, D6, and G6 (to establish G as the tonic) and eight copies of a target note. The target note was either A♭ or A in the “2” task, B♭ or B in the “3” task, C or D♭ in the “4” task, E♭ or E in the “6” task, and F or G♭ in the “7” task. On each trial, the listener strove to classify each stimulus according to its target note. Performance was best (and nearly equal) in the 2, 3, and 6 tasks, intermediate in the 4 task and worst in the 7 task. The results were well-described by a model in which a single cognitive resource controls performance in all five semitone tasks. This resource is called “scale sensitivity” here because it seems to confer general sensitivity to variations in scale in the presence of a fixed tonic.