People’s states of consciousness can be characterized by tracking language-processing brain activity, reports a paper in Nature Neuroscience. The findings could potentially be used to predict future outcomes in unresponsive patients.
Previous research has shown that brain signals could provide a signature of the loss of consciousness. Functional MRI work has reported brain responses to natural language in unresponsive patients. However, electroencephalography (EEG) is more sensitive to brain dynamics and could provide a better assessment of language processing and distinguish between disorders of consciousness, such as unresponsive wakefulness syndrome (UWS) and a minimally conscious state (MCS).
Liping Wang, Xuehai Wu and colleagues examined brain responses to speech sequences in 42 patients with MCS, 36 patients with UWS, and 47 healthy controls. They recorded EEG responses to speech stimuli with different linguistic levels (word lists, phrase tracking and sentence tracking).
They found that word processing was evident in patients and healthy controls and that phrase and sentence tracking was more apparent in patients with MCS when compared to patients with UWS. EEG data for sentence processing could be used to distinguish between MCS, UWS and control groups. The MCS group patients showed greater EEG activity in speech-tracking responses and temporal dynamics of cognitive networks while processing phrases and sentences than UWS group patients, which may help classify states of consciousness for different groups.
The authors suggest that an EEG-based model could classify individuals as a healthy control or as having MCS or UWS with high accuracy, and it could predict future outcomes of patients. However, additional work is needed to systematically compare the predictive accuracies of EEG data to current classification measures, with multiple EEG sessions from the early stage of coma to the follow-up recovery.