Objective To evaluate the critical influence of pellet density of staircase test on γ-aminobutyric acid (GABA)-dependent graft-mediated functional recovery, and to explore the transplantation of pluripotent stem cells and the role of modeling. Methods Forty-eight Sprague-Dawley rats were divided into control group, damaged group and transplanted group. The controls were not treated, the other two groups were given intracerebral injection of quinolinic Aaid (QA), and the transplantated group was injected with ventral midbrain (VM) cells from 14-day rat embryos. Three groups were divided into subgroups of 5, 12, 17, 19 weeks after transplantation. The performance of gripping and feeding of rats under high-density and low-density candy was judged through staircase test. At 28th week of transplantation, the animals in the transplanted group were assessed to estimate graft volume, the number of grafts DARPP-32 positive staining cells and damaged striatal GABAergic reinnervation case through DARPP-32 staining. Results Graft-mediated functional recovery was observed only under the high pellet condition during the bilateral and unilateral forced choice conditions. Gripping of rats was (32.9±4.5 and 47.4±2.8, P=0.024), (32.7±5.1 and 42.5±3.9, P=0.029) and (35.6±2.8 and 46.1±4.1, P=0.027; feeding of rats was (27.5±6.1 and 30.1±4.0, P=0.038), (21.6±4.6 and 29.3±2.8, P=0.031) and (26.2±3.7, 33.2±4.5, P=0.031). When tested under the low configuration, the graft provided no measurable benefit. Conclusion VM transplants can partially restore skilled forelimb sensorimotor deficits in Huntington’s disease under specific testing configuration, which can be evaluated by the stair test. Key words: Stair test; Huntington’s disease; Induced pluripotent stem cells; Transplant repair; Sensitivity