Visual, Vocal, and Verbal Markers of PTSD: A Parallel Affective Computing Approach
Abstract
Current PTSD assessment methods often compromise diagnostic precision for scalability, limiting clinical effectiveness. We propose a digital task-shifting framework that utilizes computational techniques to enhance assessment capacity while preserving clinician judgment. We implemented two structured trauma-activation tasks (trauma account and impact statement) via telehealth to elicit disorder-relevant behavioral signals, then quantified PTSD-related behaviors across modalities: verbal, vocal, and visual. Parallel unimodal statistical analyses prioritized interpretability. Comparing 49 individuals with PTSD and 44 trauma-exposed controls isolated PTSD-specific effects. Bayesian generalized linear mixed models revealed significant main effects of group across all modalities with group-by-task interaction effects specifically in verbal and vocal modalities. These findings demonstrate digital task-shifting can capture clinically meaningful variations in PTSD within a scalable framework, complementing traditional assessment methods. Strong interaction effects indicate context-dependent shifts and constrained stylistic patterns across modalities. This approach provides objective indicators of distress to potentially support clinical assessments and research via telehealth.
Related
- Associations between early antidepressant exposure, affective states, and subjective sleep quality among euthymic depressive-onset bipolar patients: a cross-sectional parallel mediation analysis
- Parental Misperception of Preschool Children’s Weight Status in China: Verbal Versus Visual Assessment and Associations with Feeding Practices
- Comparison of visual estimation of ventricular systolic function with quantitative echocardiographic markers in newborn infants.
- Psychological therapy for auditory verbal hallucinations: A qualitative assessment of voice-hearers’ experiences and preferences
- Item‐level verbal fluency psycholinguistic metrics predict preclinical Alzheimer's disease
BibTeX
@misc{yang2026visual,
title = {{Visual, Vocal, and Verbal Markers of PTSD: A Parallel Affective Computing Approach}},
author = {Yuanyuan Yang and Daiil Jun and Brett Welch and Allison Sylvia and Joel Gregory Sprunger and Jeffrey M. Girard},
howpublished = {PsyArXiv},
year = {2026},
month = sep,
doi = {10.31234/osf.io/34v5z_v1},
url = {https://osf.io/preprints/psyarxiv/34v5z_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Visual, Vocal, and Verbal Markers of PTSD: A Parallel Affective Computing Approach
AU - Yuanyuan Yang
AU - Daiil Jun
AU - Brett Welch
AU - Allison Sylvia
AU - Joel Gregory Sprunger
AU - Jeffrey M. Girard
PY - 2026
DA - 2026/09/19
PB - PsyArXiv
DO - 10.31234/osf.io/34v5z_v1
UR - https://osf.io/preprints/psyarxiv/34v5z_v1/
AB - Current PTSD assessment methods often compromise diagnostic precision for scalability, limiting clinical effectiveness. We propose a digital task-shifting framework that utilizes computational techniques to enhance assessment capacity while preserving clinician judgment. We implemented two structured trauma-activation tasks (trauma account and impact statement) via telehealth to elicit disorder-relevant behavioral signals, then quantified PTSD-related behaviors across modalities: verbal, vocal, and visual. Parallel unimodal statistical analyses prioritized interpretability. Comparing 49 individuals with PTSD and 44 trauma-exposed controls isolated PTSD-specific effects. Bayesian generalized linear mixed models revealed significant main effects of group across all modalities with group-by-task interaction effects specifically in verbal and vocal modalities. These findings demonstrate digital task-shifting can capture clinically meaningful variations in PTSD within a scalable framework, complementing traditional assessment methods. Strong interaction effects indicate context-dependent shifts and constrained stylistic patterns across modalities. This approach provides objective indicators of distress to potentially support clinical assessments and research via telehealth.
ER -