Linguistic Structure and Meaning Organize Neural Oscillations into a Content-Specific Hierarchy.

Area of Science:

• Neuroscience
• Cognitive Science
• Linguistics

Background:

• Neural oscillations are known to track linguistic information during speech comprehension.
• Previous studies often used artificial stimuli, limiting understanding of natural speech processing.
• The role of low-frequency brain activity in tracking linguistic structure in natural speech remains unclear.

Purpose of the Study:

• To investigate whether neural oscillations track linguistic structure in natural speech, independent of acoustic properties.
• To differentiate the brain's response to semantic and syntactic information from acoustic-prosodic and lexical information.
• To test neurophysiologically inspired models of language comprehension.

Main Methods:

• Measured electroencephalography (EEG) in 29 adult native Dutch speakers.
• Presented natural Dutch sentences, jabberwocky controls, word lists, and acoustically matched backward controls.
• Utilized mutual information (MI) analysis to quantify neural tracking of linguistic content.

Main Results:

• Mutual information was highest for sentences at phrasal (0.8-1.1 Hz) and lexical (1.9-2.8 Hz) timescales.
• Delta-band oscillations were modulated by lexically driven combinatorial processing, independent of prosody.
• Linguistic content (structure and meaning) organized neural oscillations beyond stimulus rhythmicity.

Conclusions:

• Neural oscillations, particularly in the delta-band, encode compositional structure and meaning during natural spoken language comprehension.
• This demonstrates that the brain infers linguistic content beyond exogenous acoustic timing.
• Findings support models where oscillations encode endogenous linguistic information during language processing.