Low-frequency cortical activity reflects context-dependent parsing of word sequences

Area of Science:

• Neuroscience
• Psycholinguistics
• Cognitive Science

Background:

• Neural activity during speech comprehension is thought to track linguistic structures.
• Debate exists whether this tracking reflects sentence-level processing or word-level features like predictability and syntax.

Purpose of the Study:

• To differentiate neural responses to sentence structure versus individual word properties.
• To investigate how context influences the parsing of word sequences into sentences.

Main Methods:

• Utilized magnetoencephalography (MEG) to record neural activity.
• Designed ambiguous word sequences parsed into different sentences depending on context.
• Analyzed low-frequency neural activity to identify context-dependent responses.

Main Results:

• Low-frequency neural activity showed a strong dependence on linguistic context.
• The difference in neural responses to the same word sequence in different contexts precisely tracked sentence-level information.
• Word predictability and syntactic properties partially explained neural responses within a context but not the context-dependent differences.

Conclusions:

• Low-frequency neural activity encodes sentence-level representations during speech listening.
• Neural responses reliably reflect how identical word sequences are interpreted as different sentences based on context.