Predictive coding of visual-auditory and motor-auditory events: An electrophysiological study

Area of Science:

• Neuroscience
• Auditory Perception
• Predictive Coding

Background:

• Auditory event-related potential (ERP) amplitude decreases for self-generated sounds, attributed to internal forward models.
• Auditory potentials are also attenuated by predictable visual motion preceding a sound.

Purpose of the Study:

• To investigate if neural underpinnings of auditory prediction are similar for motor-auditory (MA) and visual-auditory (VA) events.
• To compare the brain's response to predicted auditory stimuli originating from motor actions versus visual cues.

Main Methods:

• Utilized a stimulus omission paradigm with auditory event-related potentials (ERPs).
• Compared MA (finger tap triggering handclap sound) and VA (handclap sound with predictive video) conditions.
• Auditory stimuli were omitted in 50% or 12% of trials to elicit prediction and error signals (oN1, oN2).

Main Results:

• Auditory omissions in both MA and VA conditions induced early (oN1) and mid-latency (oN2) ERP components.
• The oN1 and oN2 components showed similar amplitude, topography, and neural sources for both MA and VA predictions.
• Prediction origins differed (motor vs. visual cortex), but the auditory processing of predicted sounds was comparable.

Conclusions:

• Motor-auditory and visual-auditory predictions activate a shared sensory template in the auditory cortex.
• This suggests a common neural mechanism for processing predicted auditory stimuli, regardless of sensory modality of the predictor.
• Findings support predictive coding theories in auditory perception and attention.