Understanding semantic impairments in schizophrenia from a predictive coding perspective

Area of Science:

• Cognitive Neuroscience
• Computational Psychiatry
• Neurolinguistics

Background:

• Schizophrenia is marked by significant semantic impairments affecting language and thought.
• Predictive coding theory posits the brain generates predictions and uses errors to update models.

Purpose of the Study:

• To provide empirical evidence supporting predictive coding as a unifying framework for schizophrenia's semantic deficits.
• To synthesize neuroimaging and cognitive data within the predictive coding model.

Main Methods:

• Review of behavioral, electrophysiological, and neuroimaging studies.
• Analysis of semantic processing in priming and fluency tasks.
• Voxel-wise modeling, graph theory, and topological analysis of semantic networks.

Main Results:

• Altered semantic processing observed in behavioral tasks.
• Reduced N400 effects indicate atypical neural responses to semantic incongruence.
• Schizophrenia exhibits disorganized semantic networks with reduced small-worldness and representational variability.

Conclusions:

• Converging evidence supports disrupted predictive mechanisms as the neurocomputational basis for semantic deficits in schizophrenia.
• Miscalibrated prediction error signaling may lead to aberrant semantic associations or impoverished processing.
• This framework offers insights into schizophrenia pathophysiology and potential therapeutic targets.