Abstract
The work loop technique was developed to assess muscle performance during cyclical length changes with phasic activation, simulating the in vivo conditions of many muscles, particularly during locomotion. To estimate muscle function in vivo, the standard approach involves subjecting a muscle to length trajectories and activation timings derived from in vivo measurements, whilst simultaneously measuring force. However, the stimulation paradigm typically used, supramaximal, 'square-wave' stimulation, does not accurately reflect the graded intensity of activation observed in vivo. While the importance of the timing and duration of stimulation within the cycle on estimates of muscle performance has long been established, the importance of graded muscle activation has not been investigated. In this study, we investigated how the activation pattern affects muscle performance by comparing square-wave, supramaximal activation with a graded in vivo activation pattern. First, we used in vivo electromyography-derived activation patterns and fibre strains from the rabbit digastric muscle during mastication and replayed them in situ. Second, we used Hill-type musculoskeletal model-derived activation patterns and fibre strains in a trotting mouse, replayed ex vivo in the soleus (SOL) and extensor digitorum longus (EDL) muscles. In the rabbit digastric muscle, square-wave activation led to an 8-fold higher estimate of net power, compared with the in vivo graded activation pattern. Similarly, in the mouse SOL and EDL, supramaximal, square-wave activation resulted in significantly greater positive and negative muscle work. These findings highlight that realistic interpretations of in vivo muscle function rely upon more accurate representations of muscle activation intensity.