An optimal control analysis of motor strategies in a brain-computer interface task

Optimality principles have provided concise and powerful descriptions of a wide variety of motor tasks, including arm-reaching, saccadic movements, and human walking. Here, we consider whether optimal control methods can be used to understand the control strategy employed in a visual brain-computer interface (BCI) task controlled by isolated cortical neurons (units). Using an inverse optimal control method we obtain, from many candidate cost functions, a weighted objective function which is consistent with average trajectories of a movement task under both brain and native limb control. The objective function has physiological and theoretical relevance, and we consider its implications for the movement strategies in each task type. Furthermore, the objective function is suggestive of a closed-loop optimal control model. The results of this paper show that optimal control theory can provide significant utility towards understanding motor strategies and system properties in BCIs.