Statistical learning in patients in the minimally conscious state

Area of Science:

• Neuroscience
• Psycholinguistics
• Cognitive Science

Background:

• Cortical activity tracks linguistic units during speech perception.
• Neural responses to linguistic units can predict outcomes in disorders of consciousness (DoC).
• Healthy individuals utilize long-term knowledge and statistical learning for linguistic processing.

Purpose of the Study:

• To investigate statistical learning in patients with minimally conscious state (MCS) and emerged minimally conscious state (EMCS).
• To assess the neural basis of statistical learning in DoC using electroencephalography (EEG).

Main Methods:

• Experiment 1: Presented real and reversed disyllabic words in isochronous sequences.
• Experiment 2: Presented trisyllabic artificial words defined by transitional probabilities.
• Utilized inter-trial phase coherence analysis and EEG to measure neural responses.

Main Results:

• Both MCS and EMCS patients showed similar word-tracking responses to real and reversed words in Experiment 1.
• MCS patients exhibited a significant word-rate EEG response to artificial words in Experiment 2, indicating statistical learning.
• These findings suggest statistical learning is possible at a minimal conscious level.

Conclusions:

• Statistical learning is present in patients with minimally conscious state (MCS).
• Residual statistical learning abilities in MCS patients offer potential for inducing neural plasticity.
• This research opens avenues for novel therapeutic strategies in DoC rehabilitation.