Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Towards the single-neuron correlates of verbal and non-verbal working memory: insights from a case study of stroke-induced aphasia
Poster Session A - Sandbox Series, Thursday, October 24, 10:00 - 11:30 am, Great Hall 3 and 4
This poster is part of the Sandbox Series.
Paolo Favero1,2, Laura Schiffl1,2, Beste Tasci1,2, Göktug Alkan1,2, Lisa M. Held1,2, Hongbiao Chen1,2, Arthur Wagner2, Bernhard Meyer2, Jens Gempt3, Simon N. Jacob1,2; 1Translational Neurotechnology Laboratory, Department of Neurosurgery, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany, 2Department of Neurosurgery, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany, 3Department of Neurosurgery, Universitätsklinikum Hamburg-Eppendorf, Hamburg, Germany
Impairments in verbal working memory have often been reported in individuals with aphasia, especially in the non-fluent variant. However, whether such impairments reflect language-specific or domain-general cognitive deficits is still debated. Are the same neuronal circuits and mechanisms responsible for storing and processing linguistic and non-linguistic information in working memory? In this study, we present a unique case of large-scale extracellular microelectrode recordings with single-neuron resolution from the prefrontal and parietal association cortex of an individual with stroke-induced non-fluent aphasia but largely intact language comprehension and domain-general cognitive functions. Neuronal data was acquired from four areas in the right hemisphere involved in working memory and semantic processing: the middle frontal gyrus, the inferior frontal gyrus, the supramarginal gyrus, and the angular gyrus. Two delayed-match-to-sample working memory tasks were designed to investigate the contribution of these regions in processing verbal and non-verbal information. In both tasks, we used non-linguistic stimuli, i.e., visually presented icons drawn from eight semantic categories, and linguistic stimuli, i.e., written or spoken words matching the pictorial stimuli. In the perceptual matching task (PM), the participant had to judge if a test stimulus was perceptually identical to a preceding sample. Therefore, semantic processing of the stimuli was not enforced. In the semantic matching task (SM), the participant judged whether the test stimulus belonged to the same semantic category as a preceding sample (e.g., “cat” and “lion”), thus enforcing semantic processing. A total of 22 sessions were performed by the participant, each consisting of two consecutive blocks (one per task). The order of the blocks was pseudo-randomized. The participant’s ceiling performance in PM demonstrated preserved non-verbal working memory, which was only mildly modulated by working memory load (decreased accuracy with increasing number of samples). The higher complexity of the SM task resulted in lower accuracy. Nevertheless, the participant reached an average accuracy of 87% for one sample (icons 93%, written words 88%, spoken words 81%) and 76% for two samples (icons 85%, written words 76%, spoken words 66%). These behavioral results suggest that semantic knowledge, lexical access, and verbal working memory are largely preserved. Ongoing analyses are directed at exploring the single-neuron and neuronal population correlates of verbal and non-verbal working memory in the right hemisphere. For example, we will investigate whether the same neuronal ensembles encode working memory content across different reference levels (iconic and symbolic) and whether these networks exhibit persistent or dynamic working memory coding. This study will generate profound insights into the neuronal foundations of human linguistic and non-linguistic cognition.
Topic Areas: Control, Selection, and Executive Processes, Disorders: Acquired