Dynamic causal modeling of the response to frequency deviants.

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Computational Neuroscience

Background:

• The mismatch negativity (MMN) is a key auditory evoked response to deviant stimuli.
• Mechanisms generating MMN, such as neural adaptation versus memory comparison, remain debated.
• Understanding MMN generation offers insights into auditory processing and predictive coding.

Purpose of the Study:

• To test hypotheses regarding the neural mechanisms underlying the mismatch negativity (MMN).
• To differentiate between models of local adaptation and memory-based comparison for MMN generation.
• To explore the role of effective connectivity in auditory evoked responses.

Main Methods:

• Dynamic Causal Modeling (DCM) applied to electrophysiological data.
• Testing biologically plausible spatiotemporal dipole models.
• Model comparison to identify the best explanation for evoked responses to deviant sounds.

Main Results:

• Dynamic causal modeling indicated that MMN generation involves changes in effective connectivity within and between cortical sources.
• Model comparison favored a model where both neural adaptation and memory comparison contribute to the response.
• Both early (N1 enhancement) and late (MMN) components are explained by the interplay of these mechanisms.

Conclusions:

• The generation of mismatch negativity (MMN) arises from a combination of neural adaptation and memory comparison processes.
• Effective connectivity changes within a hierarchical network are crucial for processing deviant auditory stimuli.
• Findings support predictive coding frameworks for hierarchical inference in the brain.