Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Differential neural organisation of laryngeal and articulatory control during speech production in Parkinson's disease: An altered auditory feedback EEG study
Poster Session C, Thursday, October 1, 10:45 am - 12:45 pm, Wangari Maathai
Miriam Huárriz1, Arpita Bose1, Fatemeh Mollaei1; 1University of Reading, United Kingdom
Parkinson’s disease (PD), the second most common neurodegenerative disorder, disrupts the forward flow of speech with symptoms including limited pitch range, monotone speech, and imprecise articulation. These symptoms seem to have a root in the way sensory information is being used and updated during speaking. The most affected speech subsystems in PD are articulatory and laryngeal control subsystems, particularly first formant frequency (F1) and fundamental frequency (f0), respectively. Behavioural studies reveal reduced response in articulatory parameters (F1; Mollaei et al., 2013) and absent or higher response in laryngeal parameters (f0; Abur et al., 2018, Mollaei et al., 2016). The aim of this research is to determine the neural signatures underlying these distinct behavioural profiles in these two subsystems. We used 64-channel electroencephalogram (EEG) during altered auditory feedback (AAF) to compare oscillatory responses to F1 and f0 within the auditory–motor network in 21 people with PD (PwPD) (12 hours off medication) and 21 age- and sex-matched controls. Participants produced /e/ vowels under conditions of AAF while their F1 (+15% and +30%) and f0 (+100 and +200 cents) were shifted upward in real time in a sustained manner (sensorimotor adaptation paradigm). Neural analyses comprised time–frequency power (Morlet wavelets, 1–80 Hz, −500 to 2000 ms) and functional connectivity (Weighted Phase Lag Index, wPLI) across delta, theta, alpha, low- and high-beta, and gamma bands within a sensorimotor ROI comprising frontal, central and parietal lobes. Group differences were evaluated with FDR-corrected linear mixed-effects models (LMMs) including age, sex, cognition, and hearing as covariates. Behavioural and neural responses were distinct by subsystem. In f0 (laryngeal control), PwPD compensated more than controls by the late epoch at +100 cents (Group × Epoch interaction, p < .001). In F1 (articulatory control), responses were comparable across groups. Time–frequency analyses showed greater parietal beta-band power in PwPD during F1 perturbation, particularly late post-onset (1500–2000 ms; high-beta, d = +0.80, q_LME = .085), but smaller effects during f0. Functional connectivity revealed two disparate Group findings in covariate-controlled LMMs: for f0, PwPD showed increased right and left frontopolar (BA10) low-beta connectivity (d = +0.76; qFDR_LME = .004), whereas for F1 they showed reduced low-beta connectivity between right inferior frontal gyrus (pars triangularis, right BA45) and right primary somatosensory cortex (right BA1) (d = -0.97; qFDR_LME = .049). Our findings show that there is a distinct pattern of behavioural and neural signatures for laryngeal versus articulatory subsystems in PD. The increased interhemispheric low-beta connectivity between frontal poles (associated with executive functions) after f0 perturbation may reflect recruitment of prefrontal regions beyond the canonical speech motor network (Guenther, 2016, p. 61), potentially supporting their enhanced compensatory responses. For F1, with no behavioural differences in adaptation, the greater parietal beta power might have offset the reduced beta connectivity between right inferior frontal gyrus and primary somatosensory cortex. These results may point to distinct neural disruptions for f0 and F1 in PD which have the potential to guide the development of subsystem-specific approaches to speech therapy.
Topic Areas: Speech Motor Control, Multisensory or Sensorimotor Integration