Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Multimodal Investigation of Word Repetition Effects: Evidence from fNIRS and EEG
Poster Session D, Saturday, September 13, 5:00 - 6:30 pm, Field House
This poster is part of the Sandbox Series.
Jinglei Ren1, Xian Zhang2, Adam Noah3, Ken Pugh4; 1Yale University
Understanding the neural mechanisms underlying the repetition effect—where previously encountered words are processed differently from novel words—is fundamental to cognitive neuroscience. In this study, we investigated the repetition effect using a multimodal neuroimaging approach, simultaneously recording EEG and fNIRS data from 30 healthy native English speakers. EEG recordings were obtained from 31 electrodes covering frontal, premotor and motor, temporal, parietal, and occipital visual regions, while fNIRS data were collected from 134 channels across similar cortical areas, providing broad spatial coverage. Traditional EEG studies offer excellent temporal resolution but limited spatial localization. By integrating fNIRS, which provides better spatial specificity through hemodynamic measurements, we aimed to achieve a more comprehensive understanding of the neural dynamics underlying repetition effects. Participants performed a semantic judgment task, deciding whether a visually presented word referred to a living animal. Each participant completed four experimental runs, each containing five blocks of six words, with one block consisting of novel words and the others repeated words from earlier trials. EEG data were preprocessed using standard pipelines in EEGLAB (Delorme & Makeig, 2004), and fNIRS data were processed using the NIRS Toolbox (Santosa et al., 2018). The results revealed significant neural differences between repeated and novel words. Event-related EEG analyses showed two distinct effects. First, novel words elicited an enhanced negativity around 170 ms compared to repeated words, corresponding to an N170 component associated with early stages of visual word form processing. Second, novel words elicited a pronounced positivity between 400 and 750 ms, consistent with the timing and polarity of a P600 component linked to semantic integration demands. Statistical analyses confirmed significant condition differences at multiple time points, highlighting the rapid and dynamic nature of repetition effects in neural processing. Complementary fNIRS analyses corroborated these findings. Activation maps revealed distinct hemodynamic patterns, with some cortical regions showing greater activation for novel words and others for repeated words. Critically, we observed a significant decrease in activation (p < 0.05) in Wernicke's area for repeated words, suggesting that repetition reduces the neural demands associated with semantic access and integration. This pattern aligns with theories of repetition suppression and enhanced neural efficiency during language processing. Together, the multimodal results demonstrate that repeated words engage partially distinct neural processes compared to novel words, reflected in both electrophysiological signatures and hemodynamic responses. The EEG findings capture both early perceptual effects and later semantic integration effects of repetition, while fNIRS provides complementary evidence of sustained regional blood flow differences. Integrating these modalities offers a more comprehensive understanding of the dynamic and distributed nature of repetition effects in the brain. In the next phase of analysis, we will correlate participants' behavioral response times with both EEG and fNIRS measures to assess how neural activity predicts performance in the semantic judgment task. This approach will help determine which modality—EEG or fNIRS—provides more precise localization of the cognitive processes supporting repetition effects.
Topic Areas: Reading, Speech Perception