Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Self-Monitoring through Rhythm: Probing Prosodic Planning with Delayed Auditory Feedback
Poster Session B, Friday, September 12, 4:30 - 6:00 pm, Field House
This poster is part of the Sandbox Series.
Ya-Ning Wu1, Gregory Hickok2; 1University of California, Irvine
Introduction: The role and integration of prosodic information in speech self-monitoring remain underexplored. Delayed auditory feedback (DAF) is a procedure where speakers hear their own voice with a time delay, which provides a valuable tool for investigating the respective roles prosodic and segmental cues. Prior study have shown that DAF around ~200 ms induces maximal stutter-like disruption in fluent speakers (MacKay, 1968). Although the effects are traditionally attributed to articulatory or phonetic-level issues, stress pattern manipulations appear to drive DAF disruption more than phonemic content (Kaspar & Rübeling, 2011). The 200 ms delay also aligns more closely with syllabic than segmental timing. Based on Hickok’s recently updated dual-stream model (2023), which describes dorsal streams that support auditory-motor coordination and timing, we would expect prosodic processing to heavily engage this dorsal stream pathway during speech production. Interestingly, Ozker et al. (2022, 2024) showed via ECoG that dorsal precentral gyrus (dPreCG) exhibits DAF-scaled error signals tightly coupled with auditory cortex. We therefore propose that prosodic frames serve as critical components for self-monitoring and motor-time alignment in speech planning. To the best of our knowledge, no study has directly attempt to isolate prosodic from segmental error sources under DAF condition with sentence level production, or examined classical envelope-tracking metrics during DAF tasks. This proposed experiment aims to address current research gap by testing prosodic contributions to speech production during auditory feedback perturbation, with future plans for neural investigation. Methods: We will use a 2×3 design comparing natural speech (normal prosody) to monotonous speech (reduced prosody) across three DAF delays: 0 ms, 100 ms, and 200 ms. Ten adult native English speakers without speech or hearing disorders will be recruited for the pilot. Speech tasks will include phonetically balanced sentences, rhythmic chants (e.g., Mother Goose), and iambic poetry. A contrastive counting task (“one two one two” vs. “ONE one ONE one”) will be employed to isolate prosodic from segmental contributions. Behavioral metrics will include disfluencies, speech rate (syllables/s), and mistimed syllables. A proposed EEG version of the study will aim to assess neural responses under identical conditions. Planned analyses include stimulus–brain envelope tracking in delta (0.5–2 Hz) and theta (4–8 Hz) bands across STG, dPreCG, and ventral articulatory areas including posterior inferior frontal gyrus (pIFG)/BA44 and Spt. Connectivity will be analyzed between STG and both (a) dPreCG and (b) the pIFG–Spt complex to test whether prosodic frame scaffolding relies more on dorsal stream mechanisms. Implications: The study will facilitate our understanding of the role of prosodic frames in temporal alignment during speech production. We predict greater DAF-induced disruption during natural prosody than monotonous speech, reflecting temporal cue tracking and mismatched feedback sensitivity. Based on Hickok's dual-stream framework, STG–dPreCG connectivity and envelope tracking in delayed conditions are expected to be impacted, especially at 200 ms delay. Examining behavioral and neural responses to DAF will have significant implications for psycholinguistic models by clarifying the role of prosodic domains, and for clinical practice by exploring the venue of temporal processing interventions for speech disorders.
Topic Areas: Speech Motor Control, Language Production