Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Cognitive and Neural Correlates Supporting Word Retrieval during Narrative Speech in Neurotypical Speakers
Poster Session A, Wednesday, September 30, 11:00 am - 1:00 pm, Wangari Maathai
Georgia Angelopoulou1,2, Dimitrios Kasselimis1,2, Dimitrios Tsolakopoulos1,2, Georgios Papageorgiou1, Georgios Tsiklis1,2, Anna Maria Psaroba1,2, Reisert Marco3, Georgios Velonakis4, Efstratios Karavassilis4, Nikolaos Kelekis4, Cornelius Weiller5, Dionysios Goutsos6, Michel Rijntjes5, Constantin Potagas1; 1Panteion University of Social and Political Sciences, 2Neuropsychology&Language Disorders Unit, 1st Neurology Department, Eginition Hospital, Faculty of Medicine, National and Kapodistrian University of Athens, 3Medical Physics, Department of Diagnostic Radiology, Faculty of Medicine, University Freiburg, 42nd Department of Radiology, General University Hospital “Attikon”, Medical School, National and Kapodistrian University of Athens, Athens, Greece, 5Department of Neurology&Neurophysiology, University Hospital Freiburg, 6Department of Linguistics, School of Philosophy, National and Kapodistrian University of Athens
Introduction Word-finding deficits are among the most salient impairments in acquired language disorders, yet less is known about lexical access in the speech output of neurotypical speakers in connected speech. Moreover, emerging evidence suggests that different word classes, such as nouns and verbs, reflect different level of cognitive demands, varying as a function of elicitation technique. Current study’s aim is to investigate the silent pauses’ frequency and duration appearing before nouns and verbs in two distinct connected speech tasks. Methods Sixty-four healthy participants (mean age: 44.58 years, SD: 12.5) were assessed in a picture description task and a personal story narration. Silent pauses’ duration and frequency and the frequency of nouns and verbs were annotated, along with silent pauses’ location with regard to nouns and verbs in both tasks. Participants underwent a neuropsychological assessment including language, memory, and executive functions’ tests. Individual row scores from neuropsychological tests were entered into a PCA with varimax rotation, using the PCA function, to reduce the number of neuropsychological scores. Structural neuroimaging data 3D T1 images – 170 slices and DWI data 64 directions were acquired for each participant. Whole brain cortical reconstruction of 3D-T1 images was conducted following the standard pipeline of FreeSurfer 7.3.2. to calculate cortical thickness in specific regions of interest of the left hemisphere. White matter fibers were reconstructed by using the global tractography approach implemented in the DTI & Fibertools toolbox. Fractional anisotropy (FA) mean values and number of streamlines were calculated and extracted for specific tracts of interest of the left hemisphere, Arcuate fasciculus, Temporofrontal Extreme capsule Fasciculus (TemEmC), Superior and Middle Longitudinal Fasciculi (SLF I and MidL). All data were analysed using the open-source statistical package R version 4.3.1. Results The PCA analysis yielded a four-factor solution (70.078% of variance) in participants’ performance. The first factor (36.72% of total variance) included neuropsychological tests assessing executive functions. The second factor (14.32% of total variance) contained indices derived from a Verbal Learning Test. The third factor (10.35% of total variance) included tasks related with access to lexical/semantic representations and the fourth factor (8.69% of total variance) included verbal short-term/working memory tasks. Linear mixed effects models revealed that frequency and duration of pauses preceding nouns were significantly higher in the picture description task compared to the personal story, whereas the frequency of pauses preceding verbs was significantly higher in the personal story. Mixed-effects models including neuropsychological factors as predictors showed that the frequency and duration of pauses preceding nouns in the picture description task were predicted by tests assessing access to lexical/semantic representations. Mixed-effects models including neuroanatomical predictors revealed that pauses before nouns in the picture description task were predicted by the integrity of the inferior longitudinal fasciculus, whereas pauses before verbs in the personal narrative were predicted by the superior longitudinal fasciculus. Conclusion Our findings suggest that word retrieval may differentiate in different elicitation techniques. Moreover, it seems to rely on distinct cognitive processes depending on lexical category but also to be supported by dissociable neural substrates.
Topic Areas: Language Production, Meaning: Lexical Semantics