Abstract
Myosin motors are fundamental biological actuators that power diverse mechanical tasks in eukaryotic cells via ATP hydrolysis. Previous work has linked myosin's velocity-dependent detachment rate to macroscopic scale muscle dynamics described by Hill's model, yet its impact on energetic flows - power consumption, output, and efficiency - remains unclear. We develop an analytical model relating myosin unbinding, quantified by a dimensionless parameter α, to energetics. Our model agrees with published in-vivo muscle data and reveals a performance-efficiency tradeoff governed by α. To experimentally validate this tradeoff, we build HillBot, a robophysical Hill muscle model that mimics nonlinearity and decouples α's concurrent effects on performance and efficiency, demonstrating that nonlinearity sensitively drives efficiency. We analyze 136 published α measurements from in-vivo muscle samples and find a distribution centered at α* = 3.85 ± 2.32. Importantly, both our analytical model and HillBot - despite operating under entirely different mechanisms - converge on the finding that this value α* of nonlinearity observed in muscle corresponds to generalist actuators that balance power and efficiency. These insights inform a nonlinear variable-impedance protocol that directly shifts along a performance-efficiency axis, which could be implemented in robotics applications.