A theoretic framework connecting speed and accuracy heterogeneity at the system and component levels

Yorie Nakahira, Caltech


The resource limitations of the brain impose severe trade-offs between speed and accuracy in neural signaling, but little theoretical foundation exists that shows how these tradeoffs impact the sensorimotor control. To develop such a foundation, we model the sensorimotor system as a control system with communication constraints and derive an analytic formula that characterizes this impact. We then show the nerve compositions that optimize the control performance for different sensorimotor control loops. This result explains the necessity of the observed nerve heterogeneity at the component level as well as the resultant performance heterogeneity at the system level, demonstrating the benefits of having a holistic perspective of both levels.