Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Speech motor profiles and their structural correlates in primary progressive aphasia
Poster Session D, Thursday, October 1, 4:30 - 6:30 pm, Wangari Maathai
Hilary Miller1, Hannah Rowe2, Frank Guenther3, Bradford Dickerson4,5; 1University of Connecticut, 2Northeastern University, 3Boston University, 4Massachusetts General Hospital, 5Harvard Medical School
Introduction: Motor speech impairments are common in primary progressive aphasia (PPA), particularly in nonfluent variant PPA (nfvPPA), where apraxia of speech is one of two defining features. The diadochokinetic (DDK) task, which requires rapid, repeated production of /pataka/, is commonly used to assess motor speech impairment perceptually across multiple domains (i.e., consistency, coordination, speed, precision, rate, accuracy). Prior work has proposed acoustic measures to quantify these domains with greater granularity. This study applied this acoustic framework to PPA to assess (1) associations between cortical thickness and acoustic measures across these domains, and (2) differences in speech motor profiles across PPA variants to inform differential diagnosis. Methods: DDK data were collected from 31 participants with PPA, remotely via Zoom. Acoustic measures were calculated from each participant’s first correct production of three repetitions of /pataka/. Four participants who were unable to complete the task were excluded from analyses. Five acoustic features were extracted from manually segmented audio using an automated Python algorithm: coordination (gap-to-syllable proportion), consistency (VOT repetition variability), speed (F2 slope), precision (between-consonant F2 variability), and rate (DDK rate). Accuracy was indexed perceptually by percent consonants correct (PCC). Structural MRI scans were available for 17 participants. Cortical thickness was extracted from Freesurfer-derived region-of-interests (ROIs) using the SpeechLabel parcellation. Brain-behavior associations were examined across a set of 24 bilateral motor speech ROIs using mass-univariate general linear models controlling for age and handedness, with Benjamin-Hochberg correction applied within each acoustic measure. Subgroup differences in speech motor profiles (i.e., the five acoustic features and one perceptual measure) were examined using a one-way MANOVA followed by univariate and pairwise contrasts. Results: Cortical thickness in the right postero-dorsal premotor cortex (pdPMC) was significantly associated with coordination (β = -0.96, partial R² = .57, p-FDR = .026), such that thicker right pdPMC was associated with greater relative pause time. Descriptively, coordination showed similar associations in right supplementary motor areas (preSMA, partial R² = .43; SMA, R² = .30). In contrast, rate, speed, and accuracy showed associations in the expected direction, with thicker cortex predicting better performance. Rate was associated with right premotor and motor regions and left preSMA; speed with left motor regions; and accuracy with right pIFs and SMA (partial R² = .25-.43). No correlates were identified for consistency or precision. Group comparisons revealed that nfvPPA differed significantly from other variants on three measures: slower speed than lvPPA (logopenic variant; d = -1.25, p = .027) and svPPA (semantic variant; d = -1.57, p = .004); reduced precision compared to svPPA (d = -1.74, p = .002); and slower rate than all groups (d = -1.81–2.55, all p < .01). Conclusion: Findings revealed a dissociation: speed, rate, and accuracy showed expected brain-behavior relationships (thicker cortex = better performance), while coordination showed an inverted relationship in right premotor cortex, where thicker cortex predicted longer pauses. This pattern may reflect compensatory right premotor recruitment supporting pause-mediated production strategies. Acoustic profiling further dissociated nfvPPA from other variants on speed, precision, and rate, supporting multidimensional acoustic measures for differential diagnosis.
Topic Areas: Speech Motor Control, Disorders: Acquired