EEG-Based Schizophrenia Detection under Subject-Wise Cross-Validation: A Multi-Method XAI Benchmark

Schizophrenia is a severe psychiatric disorder for which no objective neurophysiological biomarker is yet established for routine clinical use. Electroencephalography (EEG) offers a cost-effective and temporally precise window into cortical dynamics, yet deep learning models for EEG-based schizophrenia classification are frequently evaluated with segment-level cross-validation that allows participant-specific patterns to leak into test sets, yielding inflated accuracy estimates. This study presents a comprehensive framework combining three architecturally distinct deep learning models — EEGNet, a CNN-LSTM hybrid, and a patch-based EEG Transformer — with six explainable artificial intelligence (XAI) methods spanning gradient-based (Saliency, Integrated Gradients, DeepLIFT), game-theoretic (SHAP), activation-based (Grad-CAM), and perturbation-based (LIME) paradigms. All models are evaluated under strict subject-wise 5-fold cross-validation on the publicly available 84-participant resting-state EEG dataset from M.V. Lomonosov Moscow State University, supplemented by a multi-component regularization pipeline. The CNN-LSTM model achieved the highest classification accuracy of 0.8644 ± 0.0448 and AUC-ROC of 0.9102 ± 0.0655. Crucially, XAI analyses were conducted across all five cross-validation folds, yielding 6 × 3 × 5 = 90 model–method–fold attribution analyses. This multi-fold XAI design revealed that the right posterior temporal electrode T6 was identified as the most discriminative channel in all five folds at the grand-average level, with a grand-average importance score of 0.976. All three model architectures and five of the six XAI methods independently converged on T6 as the top-ranked channel across 5-fold averages; LIME, while identifying T6 among the most important channels, showed greater variability with F8 (right frontal) as its top-ranked channel, consistent with its perturbation-based nature. This unanimous cross-fold convergence provides robust evidence for the discriminative role of the superior temporal gyrus — a region critically implicated in auditory processing and auditory verbal hallucinations in schizophrenia.