Tracking transitional probabilities and segmenting auditory sequences are dissociable processes in adults and...

Area of Science:

• Cognitive Science
• Developmental Psychology
• Neuroscience

Background:

• Speech segmentation is challenging for infants due to the continuous nature of auditory streams.
• Transitional probability (TP) between syllables is a proposed mechanism for word discovery.
• The limits of TP-based segmentation and the role of other cues remain under investigation.

Purpose of the Study:

• To investigate the effectiveness of transitional probability (TP) for segmenting longer word units (four syllables).
• To determine the automaticity of statistical learning in neonates, even during sleep.
• To explore the role of prosodic cues in word segmentation alongside statistical learning.

Main Methods:

• Electroencephalography (EEG) was used to measure statistical learning in sleeping neonates.
• Behavioral performance in the same task was assessed in awake adults.
• Auditory streams with varying TP and prosodic cues were presented to participants.

Main Results:

• Statistical learning is automatic and effective in sleeping neonates, comparable to awake adults.
• Tracking TP alone is insufficient for segmenting longer word units.
• Subliminal prosodic cues, like pauses, significantly enhance word segmentation abilities in both neonates and adults.

Conclusions:

• Infants possess innate, adult-like abilities for speech segmentation from birth.
• Effective word segmentation relies on a combination of statistical learning and prosodic cue processing.
• A novel neural marker for monitoring learning was identified, showing increased neural response similarity post-learning.