Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Cortical Generators and Connections Underlying Phoneme Perception: a Mismatch Negativity and P300 Investigation
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster A41 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Yana Criel1, Emma Depuydt2, Marijke Miatton3, Patrick Santens3, Pieter van Mierlo2, Miet De Letter1; 1BrainComm Research Group, Department of Rehabilitation Sciences, Ghent University, Ghent, Belgium, 2Medical Imaging and Signal Processing Group, Department of Electronics and Information Systems, Ghent University, Ghent, Belgium, 3Department of Neurology, Ghent University Hospital, Ghent, Belgium
Background. The Mismatch Negativity and P300 have been widely studied in relation to the perception of pure tone contrasts. Through the administration of linguistic stimuli, these components can also reflect the processes of phoneme discrimination and phoneme categorization, respectively. In relation to pure tone processing, a network encompassing the bilateral superior auditory cortex and right frontal cortex has been described to underly the MMN. The pure tone P300, in turn, is generated by a bilateral fronto-parietal network. At present, it is unclear whether these networks also aid the perception of phoneme contrasts. From this context, the current study aimed to identify the cortical generators and functional cortico-cortical connections governing auditory phoneme perception, through the recording of the MMN and P300 during phonemic oddball tasks. Methods. In 60 healthy adults (30 males, 30 females) aged 20-80 years, electro-encephalography (EEG) was recorded from 128 electrode channels during the administration of an inattentive and attentive auditory oddball paradigm. In these tasks, the standard stimulus [be] and deviant stimulus [ge], contrasting only in terms of articulation place, were presented with an 80/20 probability. The recorded EEG data were preprocessed to extract the event-related potentials (ERPs) for the standard and deviant condition, separately. Source reconstruction of the standard and deviant ERPs was then performed for each participant using eLORETA (Pascual-Marcqui et al., 2011). We applied cluster-based non-parametric permutation testing to identify significant activation differences between both conditions in three component-specific time windows (MMN: 140-190ms, 190-240ms, 240-290ms; P300: 370-420ms, 460-510ms, 590-640ms). Functional connectivity analysis was performed between a total of 68 regions of interest, identified based on the Desikan-Killiany atlas. The maximal cross-correlation function was calculated between each ROI-pair. We applied network-based statistics (Zalesky et al., 2010) to identify significant network differences between the standard and deviant condition. Results. Activation clusters for the MMN were located in the temporal (left and right insula, left superior temporal, right temporal pole), frontal (right rostral middle frontal, pars opercularis) and parietal (left postcentral, supramarginal) cortex. Increased connectivity between the right temporoparietal and left frontal regions was found during inattentive deviant processing. P300 activation clusters were identified in the frontal (left caudal middle frontal, right precentral), parietal (right precuneus) and cingulate (bilateral posterior cingulate, right isthmus cingulate) cortex. Attentive deviant processing relied on increased intra- and interhemispheric connectivity between parietal, cingulate and occipital regions. Discussion. The present results indicate a fronto-temporo-parietal and a fronto-parieto-cingulate network to facilitate passive phoneme discrimination (MMN) and active phoneme categorization (P300), respectively. While this suggests that phoneme perception essentially draws on the same networks as pure tone processing, some activated areas, such as the supramarginal gyrus (MMN) and the insula (P300), might be specific to language processing. Both passive and active processing of phoneme contrasts showed no lateralization to the language-dominant hemisphere. The adopted stimuli, as well as the artificial way of studying phoneme processing in the current study might account for this. Alternatively, both the MMN and P300 might at least partially reflect the activation of domain-general (attention) networks, thus explaining the bilateral activation.
Topic Areas: Phonology, Methods