Understanding How Environmental Dynamics and Uncertainty Affect Adaptive Changes in Motor Planning

Abstract

The elegance and vigor of movements that organisms repeatedly display belie the incredible challenge of controlling these movements. Remarkably, the motor system can achieve smooth and robust motor control undeterred by external environmental disruptions, such as goal uncertainty or foreign dynamics. For example, goal uncertainty may be present when an individual attempts to catch a fast-approaching ball, forcing them to select where to position their arm amongst many possibilities. On the other hand, novel environmental dynamics may be present when an individual attempts to learn how to use a new tool, such as playing the piano or driving a car for the first time. To realize such fine control in the face of such environmental disturbances, the brain must rapidly adapt its motor plans based on the specific changes in the environment. Here I took three approaches to investigate the mechanisms underlying adaptive changes in motor planning that result from uncertainty and novel environmental dynamics. First, I examine how the motor system plans a movement when uncertain about the exact goal. It has been previously reported that, when faced with multiple potential targets, the brain averages the individual motor plans associated with each target, ultimately leading to a movement that is intermediate between them. However, a crucial observation of these studies is that an average of the set of individual-target motor plans often leads to a motor plan that also optimizes task performance. Thus, it is currently unclear if the motor system automatically averages potential motor plans during uncertainty, or if a single, deliberate plan is issued to optimize task performance. In the first part of this work, I demonstrate two independent experiments that systematically dissociate between motor averaging and performance-optimization. In the second part of this work, I focus on how the motor system modulates the rate of motor adaptation. While it is well known that the motor system can correct for motor errors by adapting its output, less is known of how the brain modulates the rate at which this error-dependent motor adaptation occurs. A greater understanding of this ability may offer deeper insight into the underlying organization of the motor system itself. In general, changes in net motor output are coupled with changes in limb-impedance. However, here I demonstrate that a change in net motor output itself depends on internal estimates of environmental impedance. Thus, in the second study, I show that the motor system makes movement-by-movement estimates of the impedance in the environment to modulate adaptive changes in motor output, and that this modulation can increase the amplitude of the adaptive response dramatically. Lastly, I focused on examining how error size, training duration, and the amount of adaptation influence the motor system’s ability to retain motor memories. If long-lasting benefits are to be derived from motor adaptation training, such as during rehabilitation following stroke, then retention of the performance improvements gained during practice is key. Previous work suggested that gradual, incremental introduction of novel environmental dynamics is helpful for improving retention. However, in the third part of this work, I used experimental and computational approaches to demonstrate that previously reported improvements in retention associated with gradual introductions fail to persist when other factors, including the training duration and degree of adaptation, are accounted for.