Engaging the motor system is a potent tool for improving memory, both on its own and as a component of other effective mnemonics (e.g., enactment, writing, drawing). Drawing in particular likely benefits the encoding of new memories by incorporating motor information, but also via the integration ofelaborative and pictorial information, all of which provide cues that can be used to facilitate memory retrieval. Here, we attempt to isolate the distinct influence of motor information by testingwhether passively reinstating movements during retrieval that were previously produced while encoding via drawing will boost memory performance. In an initial encoding phase, participants drew words by guiding the handle of a robotic manipulandumand pushing a button to draw. In the retrieval phase, the manipulandum instead guided the participant’s hand, chasing an on-screen bullseye. A single word appeared on-screen mid-trial, which was sometimes (half the time) one they had drawn (i.e. a ‘target'). Participants pressed a button to indicate whether they had drawn the word. During each trial, the handle moved along coordinates governed by one of three conditions: motor reinstatement (MR; the path of a drawing of the target word), motor interference (MI; a different word), or random (R; a random path). Results indicated that memory accuracy washigher and response time faster when the movements aligned with the target (MR) compared to when they did not (MI; R). This suggests that the motor system can influence memory through the reinstatement of prior movements, opening the door for future therapeutic applications.