Predictive acoustical processing in human cortical layers

Area of Science:

• Neuroscience
• Auditory Perception
• Computational Neuroscience

Background:

• Predictive processing is crucial for interpreting sensory information in dynamic environments.
• Predictive coding models describe how the brain infers and updates its understanding of the world.
• Understanding the neural basis of predictive coding requires investigating cortical architecture.

Purpose of the Study:

• To investigate the neural implementation of predictive coding in auditory perception.
• To map predictive coding processes onto the mesoscopic human cortical architecture.
• To utilize laminar-specific functional magnetic resonance imaging (fMRI) for high-resolution analysis.

Main Methods:

• Employed a cascading oddball paradigm to elicit prediction violations.
• Used laminar-specific functional magnetic resonance imaging (fMRI) to measure blood oxygenation level-dependent (BOLD) responses.
• Integrated computational modeling to analyze neural dynamics and account for signal biases.

Main Results:

• Prediction violations were hierarchically organized, activating superficial layers of the planum polare and middle layers of the lateral temporal cortex.
• Updating of the brain's internal model correlated with activity changes in deep cortical layers.
• Derived putative neural dynamics changes using a modeling approach.

Conclusions:

• The temporal cortical architecture plays a significant role in implementing predictive coding mechanisms.
• Laminar fMRI is a valuable tool for studying mesoscopic processes in extensive temporal cortical areas.
• Findings support the hierarchical nature of predictive coding in auditory perception.