Diagnosing Motor-Imagery BCI Failure: A Three-Step Diagnostic Framework

Abstract

Version 3.0.0. This version corresponds to the manuscript submitted to the Journal of Neuroscience Methods on 2 August 2026. It supersedes v2.0.0, which described a two-step framework evaluated on a single cohort. The framework now has three steps and is evaluated on three public cohorts (75 subjects). This is a preprint and has not been peer reviewed. Background. A substantial minority of users cannot achieve reliable control with motor-imagery BCIs. Existing approaches predict or treat failure, but none asks whether failure is attributable to the decoder or to the signal itself before intervention. New Method. I propose a three-step procedure. Step 1 computes classDis (Lotte and Jeunet 2018) from training-session covariance matrices. Step 2 adds a transfer gate comparing within-session and cross-session CSP-LDA performance. Step 3 applies paired permutation tests for FBCSP and separate tests for MDM, with multiple comparison correction on gate-passing subjects. Results. Pooled across three cohorts (BCI IV 2a, BNCI2015-001, Lee2019; 75 subjects), classDis correlates with cross-session decoding at r = .66, 95% CI [.50, .77]. In Lee2019, 11 subjects are flagged (20.4%), of whom four are recovered: three by FBCSP (S12, S17, S20) and one by MDM (S5), none by both. The remaining seven are transfer failures. BNCI2015-001 has one flagged subject (S11), recovered by neither decoder. BCI IV 2a has no flagged subjects. Comparison with Existing Methods. Against Blankertz et al. (2010), classDis predicts cross-session decoding rather than online feedback performance. Against Vidaurre and Blankertz (2010), classDis is classifier-free and measured before any decoder is trained. Against Ang et al. (2012), their group-level FBCSP-vs-CSP test on 2a evaluation data was non-significant at p = .059, whereas S5 is p < .001 in this framework's per-subject test. Conclusions. Failure is heterogeneous and the distinction is measurable. Any diagnostic label is relative to a specific recording and pipeline, never a claim about the person. The analysis code and every derived result table are archived separately at https://doi.org/10.5281/zenodo.21760836. The EEG data are obtained through MOABB and are not redistributed.