Decoding human mental states by whole-head EEG+fNIRS during category fluency task performance

Concurrent scalp electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS), which we refer to as EEG+fNIRS, promises greater accuracy than the individual modalities while remaining nearly as convenient as EEG. We sought to quantify the hybrid system's ability to decode mental states and compare it with its unimodal components.

We recorded from healthy volunteers taking the category fluency test and applied machine learning techniques to the data.

EEG+fNIRS's decoding accuracy was greater than that of its subsystems, partly due to the new type of neurovascular features made available by hybrid data.

Availability of an accurate and practical decoding method has potential implications for medical diagnosis, brain-computer interface design, and neuroergonomics.