Neural coding of dichotic pitches in auditory midbrain

Area of Science:

• Auditory Neuroscience
• Perception
• Psychoacoustics

Background:

• Dichotic pitches (Huggins pitch, binaural edge pitch) are perceptual illusions arising from binaural noise with changing interaural phase differences.
• These pitches are perceived tonally but lack monaural counterparts, suggesting central neural processing.
• Neural correlates of dichotic pitch at subcortical levels remain largely unexplored.

Purpose of the Study:

• To investigate neural encoding of Huggins pitch (HP) and binaural edge pitch (BEP) in the auditory midbrain.
• To identify specific neuronal response features associated with dichotic pitch perception.
• To provide a physiological basis for computational models of dichotic pitch.

Main Methods:

• Recorded single-unit responses from binaural neurons in the auditory midbrain of anesthetized cats.
• Varied stimulus frequency around each neuron's best frequency (BF) to evoke dichotic pitches.
• Utilized a Jeffress-like neural population model incorporating physiological distributions of BF and best interaural phase difference (IPD).

Main Results:

• Neuronal firing rates exhibited distinct features (peaks, troughs, edges) when stimulus frequency crossed a neuron's BF, correlating with dichotic pitch.
• These features align with models of binaural processing, including frequency tuning and sensitivity to interaural correlation.
• The neural model successfully predicted human psychophysical detection of HP when incorporating physiological neural distributions.

Conclusions:

• Demonstrates a rate-place code for dichotic pitches in the auditory midbrain.
• Establishes a physiological foundation for understanding the neural basis of pitch perception and binaural hearing.
• First study combining single-unit recordings and modeling to explore auditory neural responses to dichotic pitch stimuli.