Control Entropy: What Is It and What Does It Tell Us?

Complex tasks of motor control in humans, such as locomotion or postural control, exhibit patterns of variability that until recently have been indiscernible from random noise. Tools from the field of non-linear dynamical systems have been increasingly applied to measurements of these tasks and changes in these complex patterns have been identified. A particular tool, control entropy (CE), is a measure of the regularity, or conversely, the complexity of a signal and is used to infer the constraints present on a system. More importantly, CE can be used under nonstationary conditions, and can therefore identify changes in the complexity or constraints on a system under dynamic exercise conditions. In this review, we summarize the insight that has been gained from application of CE to signals from studies involving walking, running and postural control. We show that changing constraints can be identified during dynamic exercise and that these are reflected in changing CE. We also discuss how CE can identify increased complexity of tasks such as postural control in the fatigued state.