Perceptual chunking and its effect on memory in speech processing: ERP and behavioral evidence

Area of Science:

• Cognitive psychology
• Neuroscience
• Speech processing

Background:

• Immediate memory for spoken language is crucial for comprehension.
• Perceptual chunking, the grouping of speech into meaningful units, is a key aspect of auditory processing.
• The influence of chunk size on the memory trace of spoken items remains underexplored.

Purpose of the Study:

• To investigate the effect of perceptual chunk size on immediate memory for spoken words.
• To determine how different temporal group (TG) sizes (three vs. four items) affect memory trace quality.
• To explore the neural correlates of chunking effects using event-related potentials.

Main Methods:

• Behavioral measures and event-related potentials (ERPs), specifically N400 and P300 components, were used.
• Participants listened to monosyllabic words presented in temporal groups (TGs) of three or four items.
• Memory trace quality was assessed by analyzing N400 amplitude variations and behavioral responses.

Main Results:

• A reduced N400 amplitude was observed for items in TGs of three compared to TGs of four, indicating a better memory trace for smaller chunks.
• Behavioral responses and P300 components revealed effects related to the position of a chunk within the utterance.
• This study provides the first online measurement of perceptual chunking's effect on spoken item memory.

Conclusions:

• Perceptual chunking of speech, particularly into smaller units, enhances information buffering and memory trace quality.
• Speech is processed and stored in memory on a chunk-by-chunk basis, with chunk size being a critical factor.
• N400 and P300 ERP components effectively capture the online effects of chunking on spoken word memory.