Brain information flow analysis based on partial directed coherence on sustained attention across sensory channels

Improving sustained attention level by training across different sensory channels has a significant meaning in the fields of cognitive neuroscience. In this study, a five-day haptic training (specifically dynamic force control training) was carried out, and a visual attention task (Attention Network Task, or ANT) was used for testing the sustained attention before and after the pure haptic training. We studies not only the phenomenon but also the physiological mechanism of the improved visual attention (by ANT) with force control training (haptic only). The ERP signals recorded in the ANT tasks before and after the five-day haptic training of 14 subjects were analyzed by spectral analysis and it was shown that the power density in the frontal lobe and parietal lobe were enhanced significantly after the training. In addition, to explore the far-Transfer effect of sustained attention across different sensory channels, behavioral analysis and the information flow analysis were carried out in different brain areas. The directed brain network was constructed by using partial directed coherence (PDC) based on Granger Causality (GC). The results showed that after training, the response time (RT) and the accuracy of the ANT experiment were improved. Also, the power spectrum energy of pre-frontal lobe and parietal lobe was enhanced, and nodes in the constructed brain network in the pre-frontal lobe had more information 'flowing out' and played a more important role in the brain network after the training. Those findings showed that very likely far transfer effect exist in sustained attention training between different sensory channels such as visual and haptic channels. It may potentially provide a new way to help the treatment of people who are unable to maintain attention such as ADHD patients.