Tasha Ignatius¹, Gerome Manson¹, J. Randy Flanagan¹, Jeffrey D. Wammes¹; ¹Queen's University, Kingston, CA

Creating drawings of information can provide elaborative, pictorial, and motor cues that facilitate later retrieval from memory, but the contribution of motor information remains unclear. To test this, we had participants encode words via drawing using a robotic manipulandum. 10 mins (E1) or 1-2 days (E2) later, they completed a visual recognition task while the robotic manipulandum guided their arm through predetermined motor paths. Unbeknownst to participants, the paths were either congruent (motor reinstatement) or incongruent (interference) with their drawing of the current target word. Response time was consistently fastest with motor reinstatement, indicating that reactivating an encoded motor path made visual recognition more efficient regardless of the delay between encoding and reinstatement. However, while E1 revealed that passive reinstatement improved recognition accuracy, this benefit disappeared with the longer delay in E2. Drawings were submitted to a pretrained neural network (NN), and E1 revealed that higher-quality drawings (i.e. those more easily identified by the NN) were better recognized, regardless of reinstatement condition, but again, this pattern disappeared with the longer delay. The collective findings demonstrate that passive motor reinstatement reliably increases the speed of memory retrieval, but that any influence on accuracy is transient. Using principal components analysis (PCA) we identified two components in NN features that were associated with better memory, and these were consistent across experiments, indicating that there were predictable features that lead to more memorable drawings. Ongoing work will determine whether changes in the level of motor engagement, quality and timing of reinstatement change the observed benefit to memory. Together, these results emphasize the influence of the quality and type of encoded features as well as the timing of reinstatement in influencing memory performance.

Acknowledgements: The research was undertaken thanks in part to funding from the Connected Minds Program, supported by Canada First Research Excellence Fund, Grant #CFREF-2022-00010