Predictive musculoskeletal simulations reveal the mechanistic link between speed, posture and energetics among extant mammals

An unusual pattern among the scaling laws in nature is that the fastest animals are neither the largest, nor the smallest, but rather intermediately sized. Because of the enormous diversity in animal shape, the mechanisms underlying this have long been difficult to determine. To address this, we challenge predictive human musculoskeletal simulations, scaled in mass from the size of a mouse (0.1 kg) to the size of an elephant (2000 kg), to move as fast as possible. Our models replicate patterns observed across extant animals including: (i) an intermediate optimal body mass for speed; (ii) a reduction in the cost of transport with increasing size; and (iii) crouched postures at smaller body masses and upright postures at larger body masses. Finally, we use our models to determine the mechanical limitations of speed with size, showing larger animals may be limited by their ability to produce muscular force while smaller animals are likely limited by their ability to produce larger ground reaction forces. Despite their bipedal gait, our models replicate patterns observed across quadrupedal animals, suggesting these biological phenomena likely represent general rules and are not the result of phylogenetic or other ecological factors that typically hinder comparative studies. The fastest animals are neither the largest, nor the smallest, but rather intermediately sized, though the mechanism for this is unknown. This study built predictive musculoskeletal simulations, scaled in mass from the size of a mouse to an elephant to understand the underlying mechanisms.