Poster A50, Thursday, August 16, 10:15 am – 12:00 pm, Room 2000AB
Neuroanatomical substrates of lexical retrieval
Janina Wilmskoetter1,2, Julius Fridriksson3, Ezequiel Gleichgerrcht2, Brielle Stark3, John Delgazio2, Gregory Hickok4, Kenneth Vaden5, Argye Hillis6, Chris Rorden7, Leonardo Bonilha2;1Department of Health Sciences and Research, College of Health Professions, Medical University of South Carolina, Charleston, 2Department of Neurology, College of Medicine, Medical University of South Carolina, Charleston, 3Department of Communication Sciences and Disorders, University of South Carolina, Columbia, 4Department of Cognitive Sciences, University of California, Irvine, 5Department of Otolaryngology–Head and Neck Surgery, College of Medicine, Medical University of South Carolina, Charleston, 6Department of Neurology, Johns Hopkins University, 7Department of Psychology, University of South Carolina, Columbia
Background: Deficits in word production are commonly observed in individuals with language processing impairments due to neurological diseases, particularly post-stroke aphasia. Lexical retrieval is related to several word features such as lexical diversity, lexical sophistication, and phonological word properties, whose deficits may significantly impair communication, particularly during discourse. The neuroanatomical bases of different aspects of word production in discourse are not fully understood. This study aimed to assess the gray and white matter underpinnings related to different features of lexical retrieval during connected speech. Methods: We performed voxel-, region-of-interest-, and connectome-based lesion symptom mapping on lexical features of the words produced during discourse from 58 individuals with chronic left hemisphere stroke. In addition, we performed partial correlation on selected region-of-interests and lexical features by controlling for variance from lesion volume. Word features were obtained from picture description tasks, including measures of lexical diversity, lexical sophistication, word length / number of phonemes, phonological neighborhood density, and biphoneme probability. For transcription, we used the Computerized Language Analysis (CLAN) program, and for discourse analysis, we used open source tools (Stanford Core Natural Language Processing, Irvine Phonotactic Online Dictionary, Lexical Complexity Analyzer, Gramulator). Results: After controlling for variance from lesion volume in partial correlations, we observed that measures of lexical diversity were associated with lesions to the left inferior frontal gyrus, supramarginal gyrus, and superior temporal gyrus. Lexical sophistication was associated with lesions to the left superior and middle temporal gyrus. Lexical-phonological measures (number of phonemes, phonological neighborhood density, biphoneme probability) were associated with lesions to the left precentral and supramarginal gyri. Conclusions: Our findings indicate that lexical sophistication and phonological features of lexical retrieval during discourse dissociate to distinct lesion locations associated with the ventral and dorsal language processing stream, respectively. Measures of lexical diversity were associated with areas belonging to both the ventral and dorsal stream. Our findings contribute novel information on the neural substrates of language processing that could be used to predict lexical retrieval in discourse of individuals with chronic post-stroke aphasia and, thus, guide treatment.
Topic Area: Language Disorders