A linear oscillator model predicts dynamic temporal attention and pupillary entrainment to rhythmic patterns

Area of Science:

• Auditory neuroscience
• Cognitive psychology
• Computational modeling

Background:

• Rhythm is fundamental to music processing and influences neural activity.
• Predicting dynamic attention in complex auditory scenes remains a challenge.
• Pupillometry offers a non-invasive measure of cognitive and attentional states.

Purpose of the Study:

• To evaluate a stimulus-driven linear oscillator model for predicting dynamic attention to complex musical rhythms.
• To assess the model's ability to predict perceptual thresholds and pupillometry.
• To investigate pupil entrainment to musical rhythms.

Main Methods:

• Participants performed a deviance detection task on multi-instrument rhythmic patterns while eye-tracked.
• An adaptive thresholding algorithm adjusted deviant intensity.
• Perceptual thresholds and pupil dilation were recorded as attentional indices.

Main Results:

• The oscillator model successfully predicted perceptual thresholds, linking temporal salience to detection difficulty.
• Pupil dilation occurred for all deviants, even undetected ones, with maximum pupil size predicting detection.
• Continuous pupil signal showed entrainment to prominent periodicities in the music, demonstrating a phase relationship.

Conclusions:

• The linear oscillator model effectively predicts dynamic attention and neurophysiological signals (pupil response) in complex auditory environments.
• Pupil dynamics can be entrained by musical stimuli, offering insights into auditory attention.
• The acoustic amplitude envelope may also predict attention, similar to the oscillator model.