HHumans generally tend not to spend more energy than necessary when they perform a task. However, subjective factors, such as the comfort associated with a movement, have a significant impact on how humans behave. Some studies have used constrained optimization to explain the decision making process for movement, the effect of ergonomic factors, or even diet choice. A task’s goal represents a constraint on possible behavior, and the chosen, or optimal, behavior is determined to be that which minimizes some cost function. In biomechanics, researchers often assume costs related to energy or kinematic variability, which may miss some important subjective motivations for behavior. In this work, we leverage the optimization approach to predict and control behavior based on a more general subjective cost. However, we objectively quantify the subjective cost function in terms of mechanical work, which represents the trade-off in economy that subjective factors incur. More complete knowledge and the ability to control decisions for muscle use could benefit motor learning research, rehabilitation, and strength training.
We use an implicit approach to uncover the subjective costs associated with a number of exercise tasks. We alter task constraints, and their associated subjective costs, by unevenly weighting limb power toward a goal sum of this weighted power during exercise. The unknown subjective cost function may thus be characterized by sampling the preferred strategies for a range of different constraints. This method can be used to both characterize subjective costs of exercise in terms of objective quantities such as work, and to provide a framework with which we may create tasks which direct effort toward specific limbs.
Results indicate that healthy subjects split effort between limbs based on more than economy alone. Factors of the exercise task, such as grip type, or reach length, can alter the subject’s effort distribution toward greater use of arms or legs by about 15% of the mean net power performed. We found that implicit feedback could be used to unveil each subject’s trade-off between mechanical power generated beyond the minimum required and factors beyond economy. The implicitly weighted feedback also allowed control of the distribution of effort to allow shifts in effort toward arms or legs up to 37% of the mean net power during an exercise task. We found that the feedback could be supplied in multiple ways. We tested the use of both implicit and explicit feedback provided either through visual feedback or by changing resistance in response to the combination of implicit weights and the subject’s limb use. Subjects reduced error by 74% relative to their feedback goals and were able to perform simultaneous cognitive tasks 4.2% faster when they used implicit feedback to direct effort, rather than explicit feedback. Finally, subjective costs inform behavior outside of multi-limb exercise. In a drop landing experiment, subjects who dropped on cushioned surfaces performed up to 32% less excess work than those who landed on stiff surfaces, which allowed us to quantify the subjective cost of a more comfortable landing in terms of mechanical work.