Purpose
This study examined filled pause usage in 5-year-old English-speaking children with a history of late talking who had received intervention, compared to typically developing peers, revealing persistent distributional, temporal, and acoustic differences in filled pause usage despite intervention.
Methods
We analysed spontaneous speech samples from 73 children (36 with a late-talking history and 37 typical) using CHILDES Clinical English Ellis Weismer Corpus, focusing on filled pauses ‘um’ and ‘uh’. Filled pauses were categorised by turn-taking position and inter-pausal units, with frequency and lexical choice (uh-rate, um-rate, um-ratio) examined. Duration differences across conversational stages were also assessed. Acoustic analysis visualised pitch contours, periodic energy, and power representations. Cross-correlation analyses explored their relationships in filled pauses.
Results
The distributional analysis revealed that children with a history of late talking produced more ‘um’ in turn-initial and turn-only positions, whilst typically developing children used more in turn-medial positions. Temporal characteristics showed that children who were previously late talkers exhibited longer ‘um’ durations across all conversational stages and slower speech rates. Acoustic analysis demonstrated that their filled pauses displayed consistently higher pitch values and showed distinct cross-correlation patterns, with delayed coordination between pitch and energy compared to typically developing peers, particularly in ‘um’ production.
Conclusion
Children with a late-talking history show increased turn-initial ‘um’ use, longer durations, elevated pitch, and delayed pitch-energy synchronisation, primarily indicating persistent prosodic and processing difficulties despite receiving language intervention.
扫码关注我们
求助内容:
应助结果提醒方式:
