Work

Decision Confidence Neuron in Echo State Network for Continual Evaluation of EEG Motor Imagery Classification Quality

Year: 2026

Type: preprint

Abstract: Echo state networks (ESNs) are efficient, neuro-inspired computational frameworks well suited to time-series data. However, ESN decision confidence is typically quantified in limited ways. We propose an explicit decision-confidence readout neuron, trained from decision readout outputs, to continuously monitor confidence as decisions form. In a simulated decision task, confidence activity increased with stimulus strength, linking greater discriminability to higher confidence. We then evaluated the model on EEG-based motor imagery classification, showing that confidence activity increased with decision accuracy and discriminated correct from error decisions, particularly in higher-performing participants, reflecting human-like metacognition. Overall, this approach enables continual monitoring of decision confidence, supporting more trustworthy ESN decisions, particularly in biomedical applications.

Source: bioRxiv (Cold Spring Harbor Laboratory)

DOI: 10.64898/2026.08.31.748424

Authors: Eva Lemoine, Brendan Lenfesty, Umesh Kumar Naik Mudavath, Saugat Bhattacharyya, KongFatt Wong-Lin

Institutions: University of Ulster, Université de Poitiers, Ulster University, Derry-Londonderry Campus, Indian Institute of Technology Guwahati

Language: English

Topic: Neural Networks and Reservoir Computing

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: Royal Society, European Commission, National Natural Science Foundation of China, Medical Research Council

Awards: Uncovering the neural architecture underlying decisions abstracted from movements, G2771944745