Background <p>Stroke causes high mortality and functional loss. Stroke rehabilitation often focuses on physical independence and functional ability during activities of daily living (ADL). This multicenter randomized controlled trial compared the effect of implicit versus explicit motor training in chronic stroke rehabilitation, specifically assessing their impact on upper extremity (UE) motor function and shoulder proprioception.</p> Methods <p>Sixty-six participants were randomly allocated to the control group (CG), the implicit motor training group (IMTG), and the explicit motor training group (EMTG). Over 4&#xa0;weeks, all groups underwent 12&#xa0;one-hour sessions. The CG received conventional physiotherapy (shoulder mobilization, strengthening, and gait training), the IMTG performed dual-task exercises (e.g., reaching tasks with cognitive tasks like counting backward) to promote procedural learning, and the EMTG used tasks designed to enhance conscious movement awareness using laser pointer visual feedback. Outcome measures included the Wolf Motor Function Test, the arm motor ability test, and the laser-pointer assisted angle reproduction test.</p> Results <p>Both IMTG and EMTG showed significantly improved UE functions and ADL compared to CG (<i>p</i> ≤ 0.01). Implicit training led to faster performance in functional tasks assessed by Wolf Motor Function Test, with a&#xa0;mean difference of 5 s compared to CG (<i>p</i> &lt; 0.001), suggesting better motor skill automatization. Explicit training significantly enhanced shoulder proprioception with a&#xa0;mean decrease in the absolute angular deviation compared to CG in abduction of 1.01° (<i>p</i> = 0.01) and of 2.08° in flexion (<i>p</i> = 0.001).</p> Conclusion <p>These findings emphasize the importance of personalized rehabilitation approaches integrating both implicit and explicit strategies to optimize UE functions in people with stroke. Trial registration: ClinicalTrials.gov identifier, NCT05847595.</p>