Fifty percent of all stroke survivors remain with functional impairments of their upper limb. While there is a need to improve the effectiveness of rehabilitative training, so far no new training approach has proven to be clearly superior to conventional therapy. As training with rewarding feedback has been shown to improve motor learning in humans, it is hypothesized that rehabilitative arm training could be enhanced by rewarding feedback. In this paper, we propose a trial protocol investigating rewards in the form of performance feedback and monetary gains as ways to improve effectiveness of rehabilitative training.
This multicentric, assessor-blinded, randomized controlled trial uses the ArmeoSenso virtual reality rehabilitation system to train 74 first-ever stroke patients (< 100 days post stroke) to lift their impaired upper limb against gravity and to improve the workspace of the paretic arm. Three sensors are attached to forearm, upper arm, and trunk to track arm movements in three-dimensional space while controlling for trunk compensation. Whole-arm movements serve as input for a therapy game. The reward group (n = 37) will train with performance feedback and contingent monetary reward. The control group (n = 37) uses the same system but without monetary reward and with reduced performance feedback. Primary outcome is the change in the hand workspace in the transversal plane. Standard clinical assessments are used as secondary outcome measures.
This randomized controlled trial will be the first to directly evaluate the effect of rewarding feedback, including monetary rewards, on the recovery process of the upper limb following stroke. This could pave the way for novel types of interventions with significantly improved treatment benefits, e.g., for conditions that impair reward processing (stroke, Parkinson’s disease).
After stroke, 50% of survivors are left with impairments in arm function [1, 2], which is associated with reduced health-related quality of life . While there is evidence for a positive correlation between therapy dose and functional recovery [4, 5, 6], a higher therapy dose is challenging to implement, as it usually leads to an increase in costs commonly not covered by health insurances. However, when dose is matched, most randomized controlled trials introducing new types of rehabilitative interventions (e.g., robot-assisted therapy ) failed to show a superior effect compared to standard therapy. Thus, the need for improving therapy effectiveness remains. In search for elements of effective therapy, we hypothesize that performance feedback and monetary rewards can improve effectiveness.
It has been shown that reward enhances procedural  and motor-skill learning [9, 10] and has a positive effect on motor adaptation . Rewards mainly improve retention of motor skills and motor adaptations [9, 10, 11]. This effect was not explained by training duration (dose) as rewarded and non-rewarded groups underwent similar training schedules [8, 9, 10, 11]. In a functional magnetic resonance imaging (fMRI) study, Widmer et al. reported that adding monetary rewards after good performance leads to better consolidation and higher ventral striatum activation than knowledge of performance alone . The striatum is a key locus of reward processing , and its activity was shown to be increased by both intrinsic and extrinsic reward . Being a brain structure that receives substantial dopaminergic input from the midbrain, ventral striatal activity can be seen as a surrogate marker for dopaminergic activity in the substantia nigra/ventral tegmental area . In rodents, Hosp et al. found that dopaminergic projections from the midbrain also terminate directly in the primary motor cortex (M1) . Dopamine in M1 is necessary for long-term potentiation of certain cortico-cortical connections and successful motor-skill learning . As mechanisms of motor learning are also thought to play a role in motor recovery , rehabilitative interventions may benefit from neuroplasticity enhanced by reward.
Here, we describe a trial protocol to test the effect of enhanced feedback and reward on arm rehabilitation after stroke at matched training dose (time and intensity). We use the ArmeoSenso, a standardized virtual reality (VR)-based training system  that is delivered in two versions for two different study groups, one version with and one without reward and enhanced performance feedback. […]