Dynamic Reweighting of Auditory Modulation Filters

Area of Science:

• Auditory perception
• Psychoacoustics
• Signal processing

Background:

• Sound information is primarily conveyed through amplitude modulations (AM).
• Existing models of auditory AM processing, often bandpass filterbanks, lack definitive parameters due to varied experimental conditions and datasets.
• Uncertainty in bandwidth measurements and unresolved issues persist in auditory modulation perception research.

Purpose of the Study:

• To develop a data-driven approach for measuring auditory modulation processing.
• To characterize the auditory system's response to AM using a unified cascade framework.
• To investigate the adaptive nature of auditory modulation perception.

Main Methods:

• Employed a system identification analysis within a common 'cascade' framework.
• Utilized a single stimulus/task specification with stable rules, minimizing model constraints.
• Analyzed the relationship between trial-by-trial fluctuations in AM noise and human detection responses.

Main Results:

• Demonstrated that various auditory processing models can be integrated into a single cascade framework.
• Identified distinct elements within the auditory cascade through system identification.
• Observed adaptive changes in auditory modulation perception, shifting from lowpass to bandpass modes (Q~1) with stimulus exposure.

Conclusions:

• Auditory perception of amplitude modulations is dynamic and adaptive.
• The auditory system adjusts its filtering characteristics based on stimulus history.
• This adaptive process involves a transition from initial lowpass to broad-tuned bandpass filtering.