Temporal Flexibility of Spatial and Frequency Embedded Network Predicts Individual Learning Ability Variation in Neurofeedback Training

• Published in: IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (Online) pp. 1 - 6
• Main Authors: Liu, Shun, Wong, Chi Man, Xu, Peng, Hu, Yong, Wan, Feng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.06.2021
• Subjects: Conferences; Correlation; Frequency measurement; learning ability; multi-layer network; neurofeedback; prediction; Steady-state; Training; Virtual environments; Visualization
• ISSN: 2377-9322
• DOI: 10.1109/CIVEMSA52099.2021.9493584

Neurofeedback (NF) learning ability measured by cognitive and behavioral performance improvement after NF training shows significant individual differences. Thus, predicting an individual's future performance using before-training brain dynamics data is of great interest. Here, we introduce a novel network, which embeds spatial and frequency connectivity pat-terns to characterize the functional separation and integration ability of the brain in steady state visual evoked potentials (SSVEPs). We tested whether the flexible rewiring of this brain network can be used to predict future individual alpha band (IAB) variation, which is related to the learning ability in NF training. A total of 28 subjects underwent a two-day IAB down-regulating neurofeedback training to assess their learning ability via IAB changes. We found an as-yet-unknown significant negative correlation between the temporal flexibility of the brain network and the NF learning ability. Thus, the temporal flexibility of the brain network can serve as a predictor for the learning ability in NF training. This study will help researchers to better understand the mechanism of SSVEP and predict individual training effectiveness.