Cognitive workload: Cross-Frequency-Coupling Tensors

Gognitive workload assessment based on the tensorial treatment of EEG estimates of cross-frequency phase interactions

• Working memory (WM) related tasks are supported by networks oscillating at different frequencies. We studied cross-frequency interactions in mental arithmetic task of five cognitive workload levels (CWLs)
• Functional connectivity graphs (FCGs) were constructed, on a single trial basis, representing the phase interactions within and between Frontalθ and Parieto-Occipitalα2 neural activities.
• When single-trial FCGs were treated as tensors, the identification of CWL from the connectivity pattern became feasible with a remarkable high correct-recognition-rate (96%) .
• The proposed scheme (based on Tensor Subspace Analysis) is computational efficient and useful for real-time, personalized, applications.

For more information see:
Dimitriadis S. et al. "Cognitive Workload Assessment Based on the Tensorial Treatment of EEG Estimates of Cross-Frequency Phase Interactions",
Annals of Biomedical Engineering, 2014

You may reach the full text through our selected publications list