Characterizing multi-word speech production using event-related potentials

Area of Science:

• Neuroscience
• Psycholinguistics
• Cognitive Science

Background:

• Event-related potentials (ERPs) from electroencephalography (EEG) are valuable for studying linguistic processes.
• Most ERP studies focus on single words, not word sequences, limiting understanding of complex language production.
• Sequential word production involves distinct cognitive mechanisms like speech monitoring and planning compared to single words.

Purpose of the Study:

• To assess the feasibility of recording ERP components during word sequence production.
• To determine if individual ERP components can be isolated for each word in a sequence.
• To investigate how sequential production affects ERPs and underlying brain activity.

Main Methods:

• Scalp EEG data were collected from participants reciting word sequences from memory.
• A tongue-twister paradigm was used to elicit regular-paced word sequence production.
• Analysis focused on identifying and characterizing ERP components like error-related negativity and late left anterior negativity.

Main Results:

• A fronto-central error-related negativity (ERN) was identified for each word, sensitive to cycle and phonological similarity.
• Baseline EEG shifts indicated sustained medial-frontal activity throughout word sequences.
• A late left anterior negativity (LLAN) was observed, preceding and sustaining throughout word cycles, linked to verbal planning and execution.

Conclusions:

• ERPs are feasible for studying sequential word production, allowing isolation of components for individual words.
• Both transient ERPs and sustained EEG activity contribute to the observed brain signals during sequence production.
• This research highlights the utility of ERPs for exploring the neural dynamics of complex linguistic sequences.