Presentation

Erin Carpenter^1,3, Emily Braun¹, Manuel J. Marte^1,3, Michael Scimeca^1,3, Meryem A. Yücel^2,3, David A. Boas^2,3, Swathi Kiran^1,3; ¹Department of Speech, Language & Hearing Sciences, Boston University, USA, ²Biomedical Engineering Department, Boston University, USA, ³Boston University Neurophotonics Center, Boston University, USA

Introduction: Functional near-infrared spectroscopy (fNIRS) is a neuroimaging modality which uses optical properties of hemoglobin to measure task-related local changes in oxygenated (HbO) and deoxygenated (HbR) hemoglobin. Electroencephalography (EEG) captures neuronal activity via electrical currents. Using combined fNIRS-EEG provides an opportunity to investigate cortical activation in more naturalistic settings, as participants wear a portable cap while engaging in functional tasks (e.g., conversation, walking). Furthermore, using integrated fNIRS-EEG allows for greater specificity of task-based cortical activation than one modality alone, as it provides information on both electrical and hemodynamic activity. Specifically, in conversation tasks, EEG allows for measuring changes in frequency band power to investigate response planning and fNIRS allows for measuring changes in HbO during language production. Therefore, this study aims to use an integrated fNIRS-EEG system to measure cortical activation during conversation. Methods: Data will be collected from 15 neurotypical young adults. Participants will complete a computer-based conversation task while wearing an integrated fNIRS-EEG cap. In this paradigm, participants will either respond to conversational questions (experimental condition) or repeat sentences (control condition). To investigate the timing of response planning, the experimental condition will vary when critical information is presented in the question. In the Early-timing condition, participants will be provided with enough information to begin formulating a response halfway through the question (e.g., If you won the lottery, what would you do with the money?). In the Late-timing condition, critical information will be provided at the end of the question (e.g., What would you do with the money if you won the lottery?), and therefore participants are unlikely to begin formulating a response until hearing the entire question. Data will be collected with an integrated fNIRS-EEG cap using a NIRx NIRSport2 continuous-wave NIRS device and a LiveAmp32 EEG system. EEG time-frequency analyses will be performed over the entire question time window to investigate differences in frequency band power between the Early-timing and Late-timing conditions to determine how response planning changes as a function of when information is provided in the question. To investigate response production, differences in HbO between the experimental and control conditions during verbal responses will be evaluated. Data will be preprocessed and averaged to create six epochs (three question and three response epochs) per stimuli block. Repeated measures ANOVAs will be performed to investigate changes in HbO from baseline as the dependent variable with time (in epochs) and condition (experimental vs. control) as within-subjects variables. Anticipated Results: For response planning, we hypothesize greater decreases in alpha band power (via EEG) during the Early-timing condition as compared to the Late-timing condition in left parietal regions, indicative of a shift from attending to the question to planning a response. For response production, we anticipate greater increases in HbO (via fNIRS) from baseline during the experimental vs. control conditions in left temporal and frontal regions. Summary: Overall, the results will provide preliminary evidence on the informativeness of using an integrated fNIRS-EEG system for investigating changes in cortical activation during conversation in neurotypical adults.

Topic Areas: Language Production, Meaning: Discourse and Pragmatics