Related Experiment Video
Updated: Jun 26, 2026
Motor Imagery Brain-Computer Interface in Rehabilitation of Upper Limb Motor Dysfunction After Stroke
Published on: September 1, 2023
Cognitive-motor network integration as a behavioral marker of cognitive reserve in post-stroke rehabilitation
Katharina Kreiger1,2, Elisabeth Weiss1, Felix Fluri2,3
1Faculty of Psychology, University of Innsbruck, Innsbruck, Austria.
Background:
Functional recovery after stroke varies substantially between individuals, even after standardized inpatient rehabilitation. Cognitive reserve is increasingly considered a key determinant of recovery potential, yet it is typically approximated using indirect proxy measures that may not capture the underlying functional mechanisms of recovery. Network-based approaches may provide a more mechanistic operationalization of cognitive reserve.
Objective:
This study investigated whether (a) traditional cognitive reserve proxies predict rehabilitation response, (b) responders differ from non-responders in respect to baseline cognitive performance structure, and (c) rehabilitation response is associated with greater cross-domain cognitive-motor network integration at admission.
Methods:
In this retrospective cohort study, 100 patients (≥ 65 years) with ischemic stroke were included. Functional outcomes were assessed using a battery of motor tests at admission and discharge. A responder was defined as someone who improved in at least two functional domains. Cognitive performance was assessed using the CERAD battery. Cognitive reserve proxies included years of education and engagement in leisure activity domains. Group differences and predictors of responder status were examined using regression models controlling for stroke severity (NIHSS).
Results:
Cognitive leisure activities emerged as the strongest predictor of responder status (OR = 4.84), whereas education and other leisure domains were not retained. Exploratory factor analysis revealed two baseline cognitive dimensions (Memory, Executive-Spatial), but responders did not show significantly higher baseline cognitive scores. Network analyses demonstrated a more integrated cognitive-motor architecture in responders, characterized by higher density and lower sparsity. Dexterity and delayed verbal recall showed the highest centrality in the responder network, alongside processing speed as a consistently contributing node across centrality indices.
Conclusion:
Post-stroke cognitive reserve may be less dependent on a high cognitive performance or demographic proxies but may be due to a more integrative organization of cognitive-motor functioning. Our findings support a network-based conceptualization of cognitive reserve with direct implications for integrative rehabilitation strategies.
