Estimated 15,000 Veterans suffer a stroke each year. Stroke is a leading cause of long-term disability in the US. New strokes cost an estimated $111 million for acute inpatient care, $75 million for post-acute inpatient care, and $88 million for follow-up care in the first six months post-stroke in VHA. Yet, more than two thirds of stroke survivors have persistent hand impairment that significantly diminishes their abilities to perform activities of daily living. Dexterous manipulation of objects, such as tools, dishes, and smart phones, require not only proper hand movement, but also proper force control. Not only movement, but also control of forces from fingers has been shown to be profoundly impaired following stroke, resulting in the object being mishandled, or not handled at all, and failure at task execution. Motor control literature shows that finger movement and force control involve two independent neural controls, and therefore must be independently rehabilitated. However, conventional upper extremity therapy focuses on movement control exclusively and does not address hand force control. This gap in treatment is due to a lack of tools to provide explicit feedback on patients’ volitional finger force control. To address this gap, a novel force training tool has been developed. This tool enables stroke survivors to practice volitional finger force generation in three-dimension (3D) with explicit feedback on directional control, per best known clinical motor learning strategy. Preliminary testing of this 3D force training showed significant improvement in hand function (assessed by the Action Research Arm and Box and Block Tests, ARAT and BBT) in stroke survivors with severe hand impairment. The objective of this project is to determine if 3D finger force training is an effective tool in restoring hand function post stroke. Sixty Veterans with stroke with moderate to severe hand impairment with palpable volitional grip force will be randomly be assigned to either the experimental or control group, stratified by impairment level. Both groups will undergo 3 1-hr training sessions per week for 6 weeks. The experimental group will receive explicit feedback in 3D force, whereas the control group will receive feedback in 1D only on a computer screen. This control condition is analogous to simple squeeze ball repetitions. Training will progress by increasing influence of flexion synergy by varying posture requirements and increasing force level, introducing feedback delay, and incorporating unilateral/bilateral activity. Evaluation will occur at baseline, every 2 weeks during 6-week intervention, and at 1-month follow-up. Aim 1: Determine the effect of 3D finger force training on behavioral hand function. Hypothesis: Hand function will improve more in the experimental group than the control group. Hand function will be assessed using ARAT, BBT, and Stroke Impact Scale. Meaningfulness of the intervention will be assessed via qualitative intervie...