Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Sentence-level prosody processing in the lifespan: behavioral and neural signatures
Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai
Giada Antonicelli1,2,6, Nicola Molinaro1,3, Patricia De la Riva4, Raquel Laspiur4,5, Arantza Lopez de Turiso4, Nahia Izaguirre1,2, Simona Mancini1,3; 1Basque Center on Cognition Brain and Language, 2University of the Basque Country, 3Ikerbasque Basque Foundation for Science, 4San Sebastian University Hospital, 5Biogipuzkoa Health Research Institute, 6Fein Memory and Aging Center
Prosody is the melody of language and can convey rich meaning ranging from speaker’s intentions (speech acts) and emotions. It is still debated whether speech acts and emotional prosody are processed differently in the brain. Recent meta-analyses support a relative right-lateralization of emotional prosody, while evidence on speech acts is quite scant. Further, it is unclear how intonation processing is affected by aging, with some studies suggesting linguistic be more robust to aging than emotional prosody processing. Literature shows that aging comes with changes in auditory cortical tracking and hemispheric lateralization, but the link between those and prosody processing is largely missing. We anticipated that the access to acoustic and semantic-pragmatic subcomponents of prosody representations would be impacted differently by each of these phenomena, possibly as a function of intonation category. To test this hypothesis, we recruited 42 normal-hearing adults (28 female, age 47±19) and administered them an intonation-face expression matching task while magnetoencephalography signal was recorded. Auditory stimuli involved Neutral, Flat (no pitch variability), Question, Request, Happy and Angry prosody. For each subject and condition, we estimated (1) sensor-level cortical tracking of speech spectrogram in the delta band, (2) sensor-level intonation decoding, and (3) source activity focusing on fronto-temporal regions of interest. Age and intonation effects were tested using linear mixed-effect and beta regression models. Across analyses, age emerged as the main factor shaping prosody processing. Behaviorally, increasing age was associated with lower accuracy, although this decline was less pronounced for Question, Request, and Angry intonations. At the sensor level, age was linked to stronger bilateral delta-band cortical tracking of the speech spectrogram, together with reduced early and late decoding of intonation category over temporal sensors. At the source level, older age was associated with increased fronto-temporal activity, particularly in mid-to-late time windows. Importantly, we found no robust intonation-specific lateralization effects. In brain–behavior analyses, stronger mid-to-late right fronto-temporal activation — and, marginally, stronger cortical tracking — predicted poorer performance at older ages, whereas temporal decoding tended to support better accuracy. Together, these results suggest that prosody cognition declines with age, at a different pace across conditions. However, they do not support the classical linguistic-emotional prosody dichotomy, and the slight processing lateralization observed is mainly driven by age. Changes in auditory processing were not strong determinants for behavior in this normal hearing population. Rather, increased fronto-temporal mid-to-late activation combined with lower decoding points to a suboptimal processing strategy that favors neural noise. In this light, the interface between perception and semantic intonation representations would be the locus of vulnerability. Driven by task demands older individuals might resort to over-relying on frontal executive function to overcome the semantic integration effort signaled by increased temporal activity.
Topic Areas: Disorders: Acquired, Prosody