This systematic review sought to determine the effectiveness of mental practice (MP) on the activity limitations of the upper limb in individuals after stroke, as well as when, in whom, and how MP should be delivered. Ten electronic databases were searched from November 2009 to May 2020. Search terms included: Arm, Practice, Stroke rehabilitation, Imagination, Paresis, Recovery of function, and Stroke. Studies from a Cochrane review of MP (up to November 2009) were automatically included. The review was registered with the PROSPERO database of systematic reviews (reference no.: CRD42019126044). Randomized controlled trials of adults after stroke using MP for the upper limb were included if they compared MP to usual care, conventional therapy, or no treatment and reported activity limitations of the upper limb as outcomes. Independent screening was conducted by 2 reviewers. One reviewer extracted data using a tool based on the Template for Intervention Description and Replication. Data extraction was independently verified by a second reviewer. Quality was assessed using the PEDro tool. Fifteen studies (n=486) were included and 12 (n=328) underwent meta-analysis. MP demonstrated significant benefit on upper limb activities compared with usual treatment (standardized mean difference [SMD], 0.6; 95% confidence interval [CI], 0.32-0.88). Subgroup analyses demonstrated that MP was most effective in the first 3 months after stroke (SMD, 1.01; 95% CI, 0.53-1.50) and in individuals with the most severe upper limb deficits (weighted mean difference, 7.33; 95% CI, 0.94-13.72). This review demonstrates that MP is effective in reducing activity limitations of the upper limb after stroke, particularly in the first 3 months after stroke and in individuals with the most severe upper limb dysfunction. There was no clear pattern of the ideal dosage of MP. Copyright © 2020 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.