Hierarchical and linear sequence processing: an electrophysiological exploration of two different grammar types

Area of Science:

• Cognitive Neuroscience
• Psycholinguistics
• Computational Linguistics

Background:

• Understanding how the human brain processes complex linguistic structures is crucial for cognitive science.
• Artificial grammars provide a controlled method to study the neural basis of syntactic processing.
• Event-related brain potentials (ERPs) offer high temporal resolution for tracking neural dynamics during grammar processing.

Purpose of the Study:

• To investigate the neural mechanisms underlying the processing of finite-state grammars (FSG) and phrase structure grammars (PSG).
• To examine how the brain responds to violations within these different grammatical structures.
• To explore the role of violation position in modulating neural responses.

Main Methods:

• Utilized event-related brain potentials (ERPs) to record brain activity.
• Employed two types of artificial grammars: finite-state grammar (FSG) with local transitions ((AB)n) and phrase structure grammar (PSG) with center-embedded structures (AnBn).
• Introduced violations at early and late positions within syllable sequences of varying lengths (n=2, n=4).

Main Results:

• A posteriorly distributed early negativity was observed for FSG violations, potentially reflecting violated local expectancies.
• Both FSG and PSG violations elicited a late positivity.
• This late positivity varied with violation position in PSG, suggesting integration difficulties, but not in FSG.

Conclusions:

• The findings suggest distinct neural processing pathways for FSG and PSG.
• Early negativities in FSG processing may indicate rapid detection of local rule violations.
• The late positivity in PSG processing appears linked to the cognitive load associated with integrating complex, nested structures.