How rests and cyclic sequences influence performance in tone-scramble tasks

Abstract

When classifying major versus minor tone-scrambles (random sequences of pure tones), most listeners (70%) perform at chance while the remaining listeners perform nearly perfectly. The current study investigated whether inserting rests and cyclic sequences into the stimuli could heighten sensitivity in such tasks. In separate blocks, listeners classified tone-scramble variants as major versus minor (“3” task) or fourth versus tritone (“4” task). In three “Fast” variants, tones were played at 65 ms/tone as a continuous, random stream (“FR”), or with a rest after every fourth tone (“FRwR”), or as a repeating sequence of four tones with a rest after every fourth tone (“FCwR”). In the “Slow” variant, tones were played at 325 ms/tone in random order. In both the 3 and 4 tasks, performance was ordered from best to worst as follows: FRwR > FR > FCwR > Slow. Post hoc analysis revealed that performance was suppressed in the Slow and FCwR task-variants due to a powerful bias inclining listeners to respond “major” or “fourth” (“minor” or “tritone”) if the 4-note sequence defining the stimulus ended on a high (low) note. Overall, the results indicate that inserting regular rests into random tone sequences heightens sensitivity to musical mode.

Confirming an antiphasic bicyclic pattern of forward entrainment in signal detection: A reanalysis of Sun et al. (2021)

“…Nonetheless, even by their own measure of performance (hit rates), the finding that a segment of the population does show the effect and a segment does not is actually confirmation of the existence of the effect in a subcategory of subjects and to conclude that an absence of statistical significance at the full sample population level is evidence against existence of forward entrainment is misleading. There are several auditory phenomena that are consistently observed in a segment of normal‐hearing populations and not in others for reasons that may range from experience to genetics (Assaneo et al, 2019; Chubb et al, 2013; Ho & Chubb, 2020; Mednicoff et al, 2018).…”

Forward Entrainment: Evidence, Controversies, Constraints, and Mechanisms

“…We would, nonetheless, like to make that point that even by their own standard of using hit rates, the finding that a segment of the tested population does show the effect and a segment does not, is actually confirmation of the existence of the effect in a subcategory of subjects, and to conclude that an absence of statistical significance at the full sample population level is evidence against existence of such an effect is misleading. There are several auditory phenomena that are consistently observed in a segment of normal-hearing populations and not in others for reasons that may range from experience to genetics (Chubb et al 2013; Assaneo et al, 2019; Mednicoff et al, 2018; Ho and Chubb, 2020).…”