Listening to disfluent speech: Robust effect at processing may not extend to learning.

Area of Science:

• Psycholinguistics
• Cognitive Science
• Speech Processing

Background:

• Speech disfluencies, such as "uh" and "um", serve as predictive cues for upcoming novelty in spoken language.
• Listeners utilize these cues to efficiently process disfluent speech, leveraging predictive processing mechanisms.
• The impact of these fillers on word learning, particularly over varying time scales, remains largely unexplored.

Purpose of the Study:

• To investigate the influence of speech fillers on language processing and word learning.
• To examine the relationship between fillers, predictive processing, and prediction error.
• To determine if fillers affect word retention under different learning conditions (fluent vs. disfluent).

Main Methods:

• Five experiments were conducted involving participants exposed to novel words under fluent and disfluent conditions.
• Participants' ability to predict novelty using fillers was assessed during the exposure phase.
• Word recognition was measured at both immediate and delayed testing intervals to evaluate retention.

Main Results:

• Participants consistently used fillers to anticipate upcoming novel words during the exposure phase.
• Despite using fillers for prediction, word retention was similar across fluent and disfluent learning conditions.
• No significant difference in word retention was observed, regardless of whether participants experienced prediction errors.

Conclusions:

• Speech disfluencies effectively shape listeners' predictions during language processing.
• The predictive function of fillers does not appear to enhance long-term word retention.
• Findings suggest a dissociation between the predictive role of disfluency and its effect on learning outcomes.