統合失調症における視覚作動記憶障害の時空間的デコーディング：EEG多変量パターン解析を用いて

Deficits in visual working memory (vWM) are a fundamental cognitive characteristic of schizophrenia; however, the dynamic spatiotemporal characterization of their neural mechanisms remains unclear. The present study employs multivariate pattern classification (MVPC) and searchlight analysis to investigate neural signaling differences between patients with schizophrenia (PSZ) and healthy controls (HCS) during a vWM task. A total of 46 participants (22 PSZ, 24 HCS) completed the change detection task (1T/2T/4T). Contralateral delay activity (CDA) was extracted through ERP analysis. MVPC was employed in the temporal dimension, while a searchlight approach was employed in the spatial dimension to decode memory load (1T/2T/4T) and memory side (left/right) information. CDA amplitude was significantly lower in the PSZ group (p = .04). MVPC analysis indicated that decoding accuracy in the PSZ group was significantly lower than that in the HCS group during the 176-656 ms window (pcorrected < 0.05), suggesting reduced discriminability of multivariate ERP patterns during the delay period. Searchlight analysis revealed broadly reduced decoding across the scalp in PSZ, with the strongest group differences over posterior parieto-occipital scalp electrodes (peaking around PO3/PO4), which is consistent with prior evidence implicating posterior parietal systems in vWM maintenance. This study reveals the spatiotemporal dynamics of vWM deficits in schizophrenia using ERP decoding approaches and may inform future development of neuromarker-guided cognitive interventions.