EEG脑电解码结果受预处理方法影响极大，42%的预测会因预处理变化而翻转。
Same Brain, Different Prediction: How Preprocessing Choices Undermine EEG Decoding Reliability
- 将预处理视为可干预的反事实空间，系统分析其对解码稳定性的影响。
- 在六大数据集上，仅改变预处理方式，高达42%的试次预测结果发生翻转。
- 提出新诊断工具和正则化方法，帮助识别并缓解预处理带来的不稳定性。
脑电图（EEG）是脑机接口与临床神经科学的核心工具，但深度学习模型通常在未报告的单一预处理流程下训练与评估。本文将预处理选择形式化为反事实干预空间，发现当仅改变预处理时，跨六个数据集、四个范式，高达42%的试次级预测会发生翻转，这种变异性标准不确定性方法无法量化，因其依赖固定预处理流程。本文提供三种工具：第一，利用沃尔什-哈达玛分解分析2^7种预处理组合，揭示敏感性近似可加性，支持逐步优化；第二，引入预处理不确定性（PU），作为每试次的诊断指标，捕捉与模型置信度互补的不稳定性维度；第三，研究归一化自适应PGI（NA-PGI），一种利用预处理干预组合结构的图正则化方法，作为缓解策略，并明确其适用范围。
原文摘要 · Abstract (English)
Electroencephalography (EEG) is a cornerstone of brain-computer interfaces and clinical neuroscience, yet deep learning models are typically trained and evaluated under a single, unreported preprocessing pipeline. We formalize preprocessing choices as a counterfactual intervention space and show that EEG predictions are surprisingly unstable under this space: across six datasets spanning four paradigms, up to 42% of trial-level predictions flip when only the preprocessing changes, a variability that standard uncertainty methods do not explicitly quantify because they condition on a fixed preprocessing pipeline. We provide three tools to make this instability measurable, decomposable, and reducible. First, a Walsh-Hadamard decomposition of the 2^7 pipeline space reveals that sensitivity is near-additive in practice under the binary intervention design, enabling efficient step-by-step optimization. Second, we introduce Preprocessing Uncertainty (PU), a per-trial diagnostic that captures a dimension of instability complementary to model-based confidence. Third, we study Normalized Adaptive PGI (NA-PGI), a graph-structured regularizer that exploits the compositional structure of preprocessing interventions as one mitigation strategy with clear scope conditions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。