Spatio-Temporal Progressive Attention Model for EEG Classification in Rapid Serial Visual Presentation Task

2025-02-02

AI Summary

Addressing the challenge of detecting weak target-evoked responses in RSVP-based EEG signals, this paper proposes the Spatio-Temporal Progressive Attention Model (STPAM). STPAM employs a three-level spatial expert module to adaptively select discriminative electrodes guided by electrode topology, and a three-level temporal expert module to progressively focus on informative time segments, thereby enabling synergistic spatio-temporal feature optimization. Methodologically, it introduces the first dual-path progressive attention mechanism integrating spatial and temporal pathways, incorporating topological electrode modeling and a multi-expert collaborative architecture. Furthermore, we construct IRED—the first publicly available RSVP-EEG dataset specifically designed for weak infrared small-target detection. On IRED, STPAM achieves state-of-the-art accuracy and superior noise robustness compared to existing methods, establishing a novel paradigm for low-SNR EEG decoding.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

As a type of multi-dimensional sequential data, the spatial and temporal dependencies of electroencephalogram (EEG) signals should be further investigated. Thus, in this paper, we propose a novel spatial-temporal progressive attention model (STPAM) to improve EEG classification in rapid serial visual presentation (RSVP) tasks. STPAM first adopts three distinct spatial experts to learn the spatial topological information of brain regions progressively, which is used to minimize the interference of irrelevant brain regions. Concretely, the former expert filters out EEG electrodes in the relative brain regions to be used as prior knowledge for the next expert, ensuring that the subsequent experts gradually focus their attention on information from significant EEG electrodes. This process strengthens the effect of the important brain regions. Then, based on the above-obtained feature sequence with spatial information, three temporal experts are adopted to capture the temporal dependence by progressively assigning attention to the crucial EEG slices. Except for the above EEG classification method, in this paper, we build a novel Infrared RSVP EEG Dataset (IRED) which is based on dim infrared images with small targets for the first time, and conduct extensive experiments on it. The results show that our STPAM can achieve better performance than all the compared methods.

Problem

• EEG Signal Processing
• Temporal-Spatial Changes
• Brain Electrophysiology Analysis

Innovation

• STPAM Model
• Stepwise Attention Mechanism
• EEG Signal Analysis