To enable development of effective interventions, there is a need to complete systematic early-phase dose articulation research. This scoping review aimed to synthesize dose articulation research of behavioral motor interventions for stroke recovery. MEDLINE and EMBASE were systematically searched for dose articulation studies. Preclinical experiments and adult clinical trials were classified based on the discovery pipeline and analyzed to determine which dose dimensions were articulated (time, scheduling or intensity) and how they were investigated (unidimensional vs multidimensional approach). Reporting of dose, safety and efficacy outcomes were summarized. The intervention description, risk of bias, and quality was appraised. We included 41 studies: 3 of preclinical dose preparation (93 rodents), 2 Phase I dose ranging (21 participants), 9 Phase IIA dose screening (198 participants), and 27 Phase IIB dose finding (1879 participants). All studies adopted a unidimensional approach. Time was the most frequent dimension investigated (53%), followed by intensity (29%), and scheduling (18%). Overall, 95% studies reported an efficacy outcome; however, only 65% reported dose and 45% reported safety. Across studies, 61% were at high risk of bias, and the average percentage reporting of intervention description and quality was 61% and 67%, respectively. This review highlights a need to undertake more high-quality, early-phase studies that systematically articulate intervention doses from a multidimensional perspective in the field of behavioral motor stroke recovery. To address this gap, we need to invest in adapting early phase trial designs, especially Phase I, to support multidimensional dose articulation. Copyright © 2021 Elsevier Masson SAS. All rights reserved.