Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Adaptive short-term plasticity in semantic processing: Modality-specific and amodal networks compensate for multimodal hub disruption
Poster Session C, Thursday, October 1, 10:45 am - 12:45 pm, Wangari Maathai
Philipp Kuhnke1,2, Sandra Martin1,2, Ole Numssen2, Andrea Bruera2, Alena Gustke1,2, Mitra Morajabi1,2, Kathleen A Williams2, Markus Kiefer3, Gesa Hartwigsen1,2; 1Leipzig University, Germany, 2Max Planck Institute for Human Cognitive and Brain Sciences, 3Ulm University, Germany
Introduction Semantic knowledge is central to human cognition. Previous evidence suggests that semantic processing involves modality-specific perceptual-motor, multimodal and amodal brain regions. However, the capacity of the semantic system for adaptive short-term plasticity remains unknown. This question has direct implications for models of semantic flexibility and recovery from brain injury. Here, we asked whether modality-specific and/or amodal cortices can compensate for the disruption of a multimodal semantic hub – the left inferior parietal lobe (IPL) – by adapting their task-related activity and connectivity. Methods 40 participants received repetitive transcranial magnetic stimulation (TMS) over the multimodal IPL, or sham TMS, before performing sound and action judgment tasks on written words during fMRI. A semantic localizer (of sound and action judgments) identified the subject-specific TMS target in left IPL. TMS targeting and dosing were optimized using individual electric field simulations. Behavioral data were analyzed using generalized linear mixed models. FMRI data first underwent GLM-based univariate analysis with parametric modulation for TIME to assess plasticity-related changes across the scan. Spatial independent component analysis (ICA) on the semantic localizer then identified large-scale functional networks. Multivariate pattern analysis (MVPA) tested for TMS modulations of sound and action feature decoding within each functional network. Task-modulated functional connectivity analysis (beta-series correlation) assessed TMS effects on functional integration between large-scale networks. Results At the behavioral level, accuracy was unaffected by TMS, while response times showed a TMS x TASK x TIME interaction, indicating greater slowing of action judgments under IPL-TMS than sham. At the neural level, an analogous TMS x TASK x TIME interaction was found in left somatomotor cortex (M1/S1): Selectively during action judgments, somatomotor activity increased over time under sham stimulation, but decreased under IPL-TMS. Lower decreases in somatomotor activity were associated with less slowing after IPL-TMS, suggesting a protective role of somatomotor activity for action knowledge after multimodal hub disruption. Spatial ICA identified 21 large-scale networks engaged in semantic processing. Relative to sham, IPL-TMS selectively led to enhanced decoding of task-relevant action knowledge in three networks: SomatomotorB, Auditory, and Anterior Temporal Lobe (ATL). IPL-TMS also increased functional coupling of salience/ventral attention networks with somatomotor and auditory networks during action judgments, and with the amodal ATL network during sound judgments. Critically, these connectivity increases were associated with better behavioral accuracy in action and sound judgment tasks, respectively. Conclusion Our results suggest that modality-specific perceptual-motor and amodal networks can compensate for the disruption of a multimodal semantic hub. Upon perturbation of the multimodal IPL, somatomotor and auditory networks, as well as the amodal ATL increased their functional contributions through both activity and connectivity changes that predicted individual performance. Overall, our findings support models of semantic processing that assume a flexible neural architecture of modality-specific, multimodal and amodal cortices. Our results highlight the capacity of the semantic system for dynamic adaptation via short-term plasticity, with potential relevance for understanding resilience and recovery in semantic disorders.
Topic Areas: Meaning: Lexical Semantics,