Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Individual rhythms in solo and joint speech
Poster Session A, Friday, September 12, 11:00 am - 12:30 pm, Field House
Max Wolpert1, Nai Ding1; 1Key Laboratory for Biomedical Engineering of Ministry of Education, Zhejiang University
INTRODUCTION: Individuals exhibit stable rate preferences for motor activity, like finger tapping or music performance, and these preferences also impact auditory-motor synchronization (Scheurich, Zamm, & Palmer, 2018). We extend consideration to speech production, a highly adaptable task with layered rhythmic structure: pulse (syllabic timing) and meter (prosodic phrasing) (Fitch, 2013). To balance between these two rhythms, speakers may modulate syllabic rate to preserve prosodic timing across longer phrases (Jun 2003). METHODS: We use a self-paced and joint-reading paradigm to examine: 1. Whether individuals show stable self-paced speech rate preferences and whether prosodic structure affects these rates; 2. Whether individual rate preferences influence synchronization in joint speech. Ninety participants first completed a self-paced reading task. Based on their average syllable inter-onset interval (IOI), we invited a subset of 32 to perform synchronized joint reading. Each participant read with one rate-matched and one rate-mismatched partner. Stimuli were short sentences of varying lengths (4–6 syllables), and were read in blocks of 10 sentences to promote rhythmic entrainment. We measured speech rhythm via syllable IOIs (including lengthening and pauses) and synchronization via onset asynchrony. The difference in solo IOI rate preferences within each pair was used to predict joint asynchrony. RESULTS: Self-paced IOIs ranged from 188 to 388 ms (M = 288, SD = 50), showing substantial inter-individual variation. IOIs decreased as sequence length increased (~30 ms from 4 to 6 syllables), consistent with metrical pacing. IOIs at sentence-final positions were about twice as long as in other positions, indicating the prominence of phrase-final lengthening and pause. In joint reading, rate differences between partners did not significantly predict onset asynchrony, though pairs with closer solo rates showed a small numerical trend toward better synchronization. Regardless of initial preference, pairs tended to converge toward an intermediate rate, with one speaker leading by ~30–40 ms. Asynchrony peaked at sentence onsets and decreased over the course of the sentence, suggesting participants used both sentence-initial and sentence-final boundaries as synchronization cues. CONCLUSION: These findings suggest that while individuals have preferred speech rates, they flexibly adapt in social contexts. The weak predictive power of solo speech rate for joint synchronization highlights the flexibility of entrainment mechanisms. Compared to music, speech’s faster tempos and greater variability may make temporal alignment less reliant on individual rhythmic defaults and more dependent on adaptive cueing at prosodic boundaries.
Topic Areas: Language Production, Speech Motor Control