🤖 AI Summary
AI-enhanced electroencephalography (EEG) systems face two fundamental challenges—limited interpretability and insufficient robustness—manifesting as opaque model decisions and vulnerability to noise, artifacts, inter-subject variability, acquisition perturbations, and adversarial attacks. Method: This work presents the first systematic survey addressing both challenges jointly. We propose a unified taxonomy integrating interpretability and robustness: interpretability methods are categorized into backpropagation-based, perturbation-based, and inherently interpretable approaches; robustness mechanisms are classified according to their mitigation targets—noise/artifacts, subject variability, acquisition instability, and adversarial attacks. Contribution/Results: Leveraging cross-modal robustness analysis and methodological synthesis of explainable AI, we establish the first authoritative taxonomy for EEG-AI research. This framework exposes multi-source heterogeneous challenges and delineates concrete technical pathways and theoretical foundations for developing trustworthy brain–computer interfaces.
📝 Abstract
The close coupling of artificial intelligence (AI) and electroencephalography (EEG) has substantially advanced human-computer interaction (HCI) technologies in the AI era. Different from traditional EEG systems, the interpretability and robustness of AI-based EEG systems are becoming particularly crucial. The interpretability clarifies the inner working mechanisms of AI models and thus can gain the trust of users. The robustness reflects the AI's reliability against attacks and perturbations, which is essential for sensitive and fragile EEG signals. Thus the interpretability and robustness of AI in EEG systems have attracted increasing attention, and their research has achieved great progress recently. However, there is still no survey covering recent advances in this field. In this paper, we present the first comprehensive survey and summarize the interpretable and robust AI techniques for EEG systems. Specifically, we first propose a taxonomy of interpretability by characterizing it into three types: backpropagation, perturbation, and inherently interpretable methods. Then we classify the robustness mechanisms into four classes: noise and artifacts, human variability, data acquisition instability, and adversarial attacks. Finally, we identify several critical and unresolved challenges for interpretable and robust AI in EEG systems and further discuss their future directions.
