**Background:** An important predictor of long-term impairment after stroke is the paresis of the arm. For some years now, motor therapy of arm function is increasingly influenced by the use of virtual reality (VR). VR is intended to promote motivation and enjoyment during motor learning. Presumed is that the integration of playful elements into the virtual treatment context (gamification) ensures that patients in VR are more likely to train their motor deficits persistently and with fun. The exercises are therefore more encouraging and motivating than in a conventional setting. However, it is a fact that the myth of increased motivation through VR scientifically has not been proven.
**Objectives:** 1) To discuss the influence of motivation through VR in the context of rehabilitation and with regard to the effects of motor learning and 2) to identify available clinical studies that investigate motivation as an endpoint in group comparison (VR vs. conventional therapy) in order to evaluate it qualitatively and quantitatively.
**Methods:** Systematic literature search in the databases Medline (Pubmed), PEDro and Cochrane Library. Study results were summarized narratively and evaluated using meta-analysis methods.
**Results:** The qualitative analysis included N=10, the quantitative N=6 studies of moderate to low quality, PEDro score 5.3(±1.2). Study sample sizes were generally small (21 ±12) and interventions varied in terms of goals, treatments, measurements, and virtual reality devices used. Control groups usually received therapy based on a standard‐care approach. The influence of VR on motivation was not statistically significant: SMD 0.02 (CI -0.27 - 0.31; p 0.9; I2 = 0%, 185 Participants).
**Discussion:** The results suggest that the use of VR is not more motivating than conventional therapy approaches to improve upper extremity function after stroke. Neither types of measuring instruments, nor time since stroke onset or used VR systems and comparative therapies had a significant influence on the results. The quality of the evidence for the outcome is limited due to small sample sizes, inconsistent results and partly selective reporting.
**Conclusion:** Results need to be seen as preliminary and do not support evidence about potential long-term effects of the therapy. Larger phase II studies are necessary, in which the potential effects of VR on motivation are systematically investigated by identifying relevant outcomes and applying appropriate inferential statistical analyses.
**Keywords:** virtual reality, stroke, rehabilitation, motivation
