Balancing Performance and Stability in Motor Imagery EEG Decoding: A Linear-First Approach

Abstract

Motor Imagery (MI) EEG decoding lacks a principled way to decide when a complex nonlinear classifier is warranted over strong, data-efficient linear baselines. While literature often chases marginal accuracy gains, it rarely balances these against computational burden or stability under neurophysiological non-stationarity.

This work operationalizes a “Linear-First” rule using statistical inference and drift-aware neuroscience. We argue that a complex alternative should only be chosen if the evidence shows it is better than a simpler baseline; otherwise, the linear option is kept for the sake of interpretability and efficiency.

Methodological Framework

1. The Linear-First Decision Rule

We evaluate feature-representation and fixed-feature classifier choices under nested cross-validation (5 outer and 3 inner folds, as in dissertation Chapter 1), with separate measurements of predictive accuracy and computational cost.

• Protocol: Every learnable step (scaling, subspaces, shrinkage) is fit strictly inside the inner loop to prevent data leakage.
• Logic: The linear candidate family (LDA, logistic regression, linear SVM, elastic net) is compared against a pre-specified RBF-SVM, which replaces the linear family only when the evidence supports the change.

2. Computational Pareto Fronts

To ensure fair comparison, we construct Pareto fronts in two dimensions to adjudicate near-ties between models:

• Accuracy vs. Latency: Measured using a single-core inference protocol.
• Accuracy vs. FLOPs: Auditing budget alignment to ensure gains are not artifacts of increased search effort.

3. Neurophysiological Drift Analysis

We quantify within-session non-stationarity to connect decoding performance to sensorimotor physiology:

• Spectral Drift: Phase-wise changes in $\mu/\beta$ band power (ERD/ERS carrier).
• Geometric Drift: Affine-invariant distances on the Symmetric Positive Definite (SPD) manifold between phase-wise covariance means.

Key Insights

The analysis examines stability and computational cost alongside decoding accuracy. These research comparisons do not establish clinical performance or validate a closed-loop rehabilitation system.

For inquiries regarding this research, please contact the author directly.