Meeting Abstract
Maximum speed and peak acceleration are important performance traits for understanding variation in the evolution of morphology, biomechanics, ecology and behavior. Many animals rely on quick bursts of locomotion when escaping predators and capturing prey. Hence strong selective pressures can occur on these traits. A variety of methods can be used to measure maximum running speed and, in general, we have a high degree of confidence in our ability to accurately and repeatedly estimate maximum speed. Estimation of peak acceleration is frequently achieved via high speed video and can be quite inaccurate depending on the details of video capture and the method used to smooth and compute the second derivative of the displacement data. Other methods, such as force plates and accelerometers, allow for more direct estimates of acceleration but have various limitations. We argue that we can avoid estimating peak acceleration, and the issues associated with its estimation, by redefining acceleration performance. A burst locomotor event is a movement that starts from a standstill and involves an accelerative phase up to a maximum speed. Limbed animals achieve such movement by cycling their limbs wherein each footfall results in a substrate reaction force that causes the animal to accelerate forward and gain speed. In this context, it is the speed gained per step and the number of steps that principally determine maximum speed. Acceleration performance can thus be defined as the per step gain in speed, which can accurately be estimated by overlaying footfalls onto the velocity curve. This definition of acceleration performance avoids the errors associated with computing acceleration from video-based displacement data and is linked to how step-based accelerations generate maximum running speed.
