Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Sensory-motor mechanisms for verbal working memory
Poster Session C, Saturday, September 13, 11:00 am - 12:30 pm, Field House
B. Liang1, A.M. Earle-Richardson1, D. Southwell2,3,4,6, G. Grant3,6, M. Zafar1,5,6, B. Frauscher1,2,6, G. Hickok9, G. B. Cogan1,2,3,6,7,8; 1Department of Neurology, Duke University, 2Department of Biomedical Engineering, Duke University, 3Department of Neurosurgery, Duke University, 4Department of Neurobiology, Duke University, 5Department of Pediatrics, Duke University, 6Duke Comprehensive Epilepsy Center, Duke University, 7Center for Cognitive Neuroscience, Duke University, 8Psychology and Neuroscience, Duke University, 9Departments of Cognitive Sciences and Language Science, University of California, Irvine
Conversational language relies on the transient maintenance of verbal working memory (vWM). Yet the precise neural mechanisms supporting vWM remain elusive. To address this gap, in this study, we recorded and analyzed neural signals from patients implanted with stereoelectroencephalography (sEEG) electrodes performing a lexical repetition task with a vWM delay. 37 patients (mean age = 31, 21 female) with electrodes implanted as part of their surgical epilepsy evaluation were recruited in this study. In each trial, patients heard one item from a corpus of 84 CVCVC-structured words or 84 matched nonwords (e.g., “bacon,” “valuk”), and repeated the item aloud after a visual Go cue following a randomized 1.125±0.125s delay. As a control analysis, 18 of the recruited patients completed an additional lexical repeat task without a delay (immediate repetition). To characterize the spatiotemporal dynamics of vWM, we first preprocessed the intracranial recordings and identified clusters of local neural engagement as significant increases in high-gamma power (HG, 70–150 Hz) relative to a pre-trial baseline (cluster-based correction, p < .05). Electrodes with significant HG responses during vWM delay (Delay electrodes) were found (N = 1009) distributed in the dorsal and ventral speech processing streams (Hickok & Poeppel, 2007). Within these significant delay electrodes, 33.11% (N = 402) also had auditory responses (Auditory delay electrodes), 20.51% (N = 249) had motor responses (Motor delay electrodes), and 38.06% (N = 462) had both auditory and motor responses (Sensory-motor delay electrodes), spatially adjacent to both Auditory and Motor electrodes, replicating previous results (Cogan et al., 2014, 2017). Together, this shows that vWM is supported largely by neural mechanisms for sensory-motor speech processing. Analyses on the subset of patients with the control task demonstrated that both Delay electrodes with auditory and motor responses functioned similarly even without delay (Auditory + Sensory-motor: 79.22% overlapped the corresponding category in control; Sensory-motor + Motor: 67.67% overlapped). This control analysis indicates that the vWM neural responses stem from sensory-motor mechanisms in general speech processing, even when no vWM component is explicitly required. To assess the representational content of vWM, we independently performed a general linear modeling (GLM) analysis of hierarchical linguistic features (acoustic, phonemic, and lexical status for word vs. nonword) on high-gamma neural responses in electrodes during the delay period (cluster-based correction, p < .05). We identified electrodes that were significantly predicted by higher-level phonemic (N = 451) and lexical-status (N = 309) features in delay, but the encoding of acoustic features was largely absent (N = 44), suggesting that vWM represents higher-level phonological and lexical features of verbal stimuli. Taken together, we found that vWM is supported by sensory-motor speech processing neural mechanisms that maintain higher-level linguistic representations.
Topic Areas: Multisensory or Sensorimotor Integration, Control, Selection, and Executive Processes