Age-Related Changes in Action Observation EEG Response and Its Effect on BCI Performance

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Science

Background:

• Action observation-based brain-computer interfaces (AO-BCI) leverage both visual evoked potentials and sensorimotor rhythms for neurorehabilitation.
• Current AO-BCI research predominantly targets younger populations, leaving age-related effects under-explored.

Purpose of the Study:

• To investigate age-related differences in electroencephalography (EEG) responses during AO-BCI tasks.
• To analyze how prefrontal EEG features and whole-brain functional networks differ between older and younger adults using AO-BCI.
• To establish quantitative relationships between EEG characteristics and AO-BCI classification accuracy across age groups.

Main Methods:

• Comparative study involving 18 older and 18 younger subjects performing an AO-BCI task.
• Task discriminant component analysis (TDCA) for decoding observed actions.
• Analysis of prefrontal EEG approximate entropy (ApEn), sample entropy (SamEn), rhythm power ratios (RPR), and whole-brain functional network connectivity (in alpha, beta, and theta bands).
• Regression analyses to correlate EEG features with classification accuracy.

Main Results:

• Older adults exhibited significantly lower TDCA accuracy (77.01% ± 14.67%) compared to younger adults (87.22% ± 15.22%).
• Distinct age-dependent patterns were observed in prefrontal ApEn, SamEn, and RPR.
• Significant intergroup differences in alpha and beta band connectivity strength, and altered theta band network topology (reduced prefrontal nodal degree, enhanced global efficiency) in older adults.
• A significant inverse relationship was found between beta/theta RPR and overall accuracy; beta/theta RPR and beta band ApEn were identified as key factors influencing individual differences in accuracy.

Conclusions:

• Age significantly impacts AO-BCI performance, with older adults showing reduced classification accuracy.
• Specific EEG features, including prefrontal entropy, rhythm power ratios, and brain network connectivity patterns, exhibit age-dependent alterations.
• These findings provide crucial insights into the neurophysiological mechanisms underlying age-related differences in AO-BCI and can guide the optimization of BCI technology for neurorehabilitation across diverse age groups.