Rehabilitation robots hold significant promise for maximizing recovery of lost function caused by neurological impairments such as stroke. However, to date, robots have not produced functional gains that are markedly better than those achieved by traditional therapy. While rehabilitation robots can be used to deliver precise movement prescriptions, they do not provide a means for designing treatments that target the neural control deficits of a particular patient. In contrast, neuromusculoskeletal models have recently advanced to the point of being able to design personalized treatments that target an individual’s neural control deficits, but models provide no way of implementing treatments. These two observations lead to “The Reese’s Peanut Butter Cup Effect,” where combining two traditionally separate fields – rehabilitation robotics and neuromusculoskeletal modeling - has the potential to produce better outcomes than what either field could achieve separately. Muscle synergies are the proposed common element for tying these two fields together and facilitating a personalized treatment design process.