Tone-in-noise detection using envelope cues: comparison of signal-processing-based and physiological models

Area of Science:

• Auditory Neuroscience
• Psychoacoustics
• Signal Processing

Background:

• Tone-in-noise detection relies on auditory cues like energy, envelope, and fine-structure.
• Previous research linked these cues to listener performance in detecting tones in noise.
• Envelope cues are crucial for understanding auditory signal detection.

Purpose of the Study:

• To investigate the role of envelope cues in diotic and dichotic tone-in-noise detection.
• To compare the predictive power of stimulus-based versus physiological models of auditory processing.
• To determine if physiological models, including neural nonlinearities, improve predictions of human performance.

Main Methods:

• Examined envelope cues using stimulus-based models (envelope slope, binaural slope of interaural envelope difference).
• Applied physiological models simulating the auditory nerve (AN), cochlear nucleus, and inferior colliculus (IC).
• Incorporated neural mechanisms like modulation gain and nonlinear transformations in physiological models.

Main Results:

• Stimulus-based models do not account for crucial nonlinearities in auditory processing.
• Physiological models, including the AN and cochlear nucleus, incorporated modulation gain.
• A model inferior colliculus (IC) cell's response fluctuations predicted human performance across noise maskers, similar to previous stimulus-based models.

Conclusions:

• Physiological models that include neural mechanisms affecting stimulus envelope encoding can predict listener performance in tone-in-noise detection.
• These models offer a more comprehensive understanding of auditory processing compared to stimulus-based approaches.
• Neural nonlinearities play a significant role in how the auditory system encodes envelopes for sound detection.