Schedule of Events | Search Abstracts | Invited Symposia | Symposia | Poster Sessions | Data Blitz

Brain vocabulary: EEG-based semantic feature maps for verbs

Poster Session F - Tuesday, March 10, 2026, 8:00 – 10:00 am PDT, Fairview/Kitsilano Ballroom

Milena R. Osterloh^1,2 (), Laura A. Ciaccio^1,3, Johann Berger¹, Luigi Grisoni^1,4, Friedemann Pulvermüller^1,2,5,6; ¹Brain Language Laboratory, Freie Universität Berlin, Berlin, Germany, ²Einstein Center for Neurosciences Berlin, Berlin, Germany, ³Department of Brain and Behavioral Sciences, University of Pavia, Pavia, Italy, ⁴Center for Translational Neurophysiology of Speech and Communication, Fondazione Istituto Italiano di Tecnologia, Ferrara, Italy, ⁵Berlin School of Mind and Brain, Humboldt Universität zu Berlin, Berlin, Germany, ⁶Cluster of Excellence ‘Matters of Activity’, Humboldt Universität zu Berlin, Berlin, Germany

Neuroimaging studies show that different semantic categories, and even individual words, elicit distinct patterns of brain activity (e.g., Antoine et al., 2024; Grisoni et al., 2021). Furthermore, recent work demonstrates that semantically similar words further evoke more similar patterns of neural activity (e.g., Carota et al., 2023; Fernandino et al., 2022). However, previous research has focused on concrete nouns, leaving verbs and abstract domains relatively unexplored. This EEG study investigates how word-specific spatiotemporal patterns of brain activity reflect the semantic features of verbs across concrete and abstract categories. Fifty-two healthy native German speakers participated. Stimuli comprised 120 German verbs from four semantic categories: Actions-Face, Actions-Hand, Emotion, and Performative. Verbs were presented in a passive reading paradigm. A collapsed localizer approach identified time windows of interest in the grand-average ERP. Mixed-effects linear regression models were applied to each time window to assess category-related effects. ERP differences among semantic categories emerged around 200 ms post-stimulus. A significant interaction between semantic category and topographic gradient was observed in time windows centered around 250, 300, and 400 ms, with more positive amplitudes for Actions-Hand verbs at posterior sites. Concreteness effects emerged from ~200 ms, showing a significant interaction between concreteness and gradient within the same time windows. These findings indicate that neurophysiological differences between semantic verb categories emerge from ~200 ms post-onset, with pronounced effects for Actions-Hand verbs. Importantly, early ERP responses also reflect abstract-concrete distinctions. Ongoing work applies RSA to compare grounded semantic descriptions (35 rated dimensions) with EEG responses to single words.

Topic Area: LANGUAGE: Semantic