Speech acoustics, prefrontal activation, and their coupling during affective prosody production in dementia of the Alzheimer’s type

Introduction Affective prosody, the expression of emotion through speech, is often disrupted in dementia of the Alzheimer’s type (DAT), yet its acoustic and neural bases remain poorly understood. Extending our earlier single-group investigation, we compared individuals with DAT and cognitively healthy older adults across speech acoustics, prefrontal activation, and their coupling. Methods Ten individuals with DAT and ten cognitively healthy older adults of similar mean age produced spontaneous happy, sad, and neutral speech while prefrontal cortical activity was recorded using functional near-infrared spectroscopy (fNIRS). Acoustic features were extracted and analyzed alongside cortical activation and speech–brain activation coupling. Results The groups differed reliably in the formant domain, while intensity, fundamental frequency, and formant structure varied by affective condition. Overall prefrontal activation did not differ between groups, but group-by-emotion interactions revealed opposite response directions: healthy adults generally showed increased activation during affective speech, whereas individuals with DAT showed decreased activation. Seven of eight contrasts survived correction for multiple comparisons. The pattern was replicated in total hemoglobin and remained robust to the removal of individual participants. Speech–brain activation coupling also differed between groups, with large effect sizes (Cohen’s q = 0.64–0.66), although these exploratory effects did not survive correction for multiple comparisons. Discussion DAT was characterized not by uniform reduction but by reorganization of the relationship between affective speech and its cortical support. Joint analysis of acoustic and neural signals may provide a more comprehensive understanding of affective communication in dementia, while the exploratory coupling findings require confirmation in a larger sample.