cs.LGOct 2, 2026

Articulatory Entrainment and Coordination Complexity in Spontaneous Autistic and Non-autistic Dialogue

Authors: Thanushi Withanage, Carol Espy-Wilson, Elizabeth Redcay, Desi Jones, Noah Sasson

Organizations: Department of Electrical and Computer Engineering, University of Maryland College Park, USA · Department of Psychology, University of Maryland College Park, USA · Department of Psychology, The University of Texas at Dallas, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Articulatory entrainment, the adaptation of vocal tract coordination to facilitate interaction remains underexplored in spontaneous dialogue, particularly among autistic speakers. Many prior studies have utilized task-based, phoneme-level analyses with invasive measurement techniques. Here, we introduce a speaker-independent framework to quantify articulatory entrainment in spontaneous dyadic conversations among autistic and non-autistic adults using acoustic-to-articulatory inversion and coordination complexity metrics. We examine temporal changes in articulatory coordination across interaction. Non-autistic dyads exhibit increasing coordination complexity and stronger entrainment over time, autistic dyads show moderate effects, and mixed dyads demonstrate the least alignment. Greater articulatory entrainment correlates with higher self-reported conversational success, indicating increase in coordination complexity as a marker of effective social interaction.

Figures & tables

Figure 1: Coordination simulation and eigen spectra comparison for simple, complex and erratic speech
Figure 2: Eigen spectra for an individual for each quarter of the conversation
Figure 3: Difference eigen spectra for an individual for each quarter of the conversation with first quarter eigen spectra as the reference
Figure 4: (a) Mean & standard deviation variations of ω\omega of NA-NA dyads vs. quarter of conversation for the CANDOR dataset. (b) Percentage of Conversations showing positive entrainment ( e>0e>0 ) and high PCS in NA-NA dyads of CANDOR dataset. (c) Mean & standard deviation variations of ω(h)\omega^{(h)} vs. two halves of conversations for different dyad groups (by speaker) in the UTD dataset. (d) Entrainment vs. dyad type in UTD dataset.
More in this topic

Autism Spectrum Disorder

Autism Spectrum Disorder research studies its developmental, behavioral, and neural features and ways to improve assessment and support.

See all 37 papers

Explore similar work

CardsList

  1. Acoustic and Facial Markers of Perceived Conversational Success in Spontaneous Speech

    Mar 3, 2026Thanushi Withanage, Elizabeth Redcay, Carol Espy-Wilson

    Department of Electrical and Computer Engineering, University of Maryland College Park, MD, USA · Department of Psychology, University of Maryland College Park, MD, USA

  2. TRACE: Temporal Relationship-Aware Conversational Entrainment Detection in Dyadic Speech

    Jun 29, 2026Sathvik Manikantan Napa Ugandhar, Hao Zhang, Alison Gunzler +5Emotion Recognition in Conversations

    Department of Computer Science, Johns Hopkins University, USA · Department of Electrical and Computer Engineering, Johns Hopkins University, USA · Amazon Research, UK +1

  3. ART-NAD: An Articulatory Inversion-based Neural Acoustic Distance for Pathological Speech Intelligibility Assessment

    Sep 21, 2026Bence Mark Halpern, Thomas Tienkamp, Defne Abur +1Speech Quality AssessmentSpeech Processing

    Nagoya University Nagoya, Japan · University of Cologne Cologne, Germany · University of Groningen Groningen, the Netherlands