The Same Motor Behavior with a Different Theoretical Explanation

Although the types of movement skills selected for study by researchers aligned with a hierarchical

view of motor control are quite different from those studied by advocates of dynamical approaches to motor control, one movement pattern that has been studied by both groups is human locomotion. Specifically, both groups of researchers have studied the transition that occurs between walking and running as gait velocity increases to provide evidence in support of

certain theoretical assumptions associated with their respective theories. Not surprisingly, both

groups of researchers have explained the walkto-run transition very differently. Shapiro, Zernicke, Gregor, and Diestel (1981) studied the gait cycle at different speeds for the purpose of testing the hypothesis that relative timing between limbs constitutes an invariant parameter

of a generalized motor program (GMP) used to control a class of movements. They further hypothesized that, because walking and running belong to different classes of movement, the relative timing would differ between the two movement patterns. The relative timing associated with each of the four phases of the gait cycle plotted for the right limb (i.e., heelstrike and midstance, midstance to toe-off, toe-off to midswing, and midswing to heelstrike) was invariant. That is, the duration of each phase of the step cycle expressed as a percentage of the total step cycle was similar for walking speeds that ranged between 3 and 6 mph. Once the speed of the treadmill on which the subjects were initially walking was increased to 8 mph, the relative timing characteristics of the step cycle changed as the subject transitioned to running. The relative timing of the four phases did not vary, however, for running speeds ranging between 8 and 12 mph. On the basis of their findings the authors not only concluded that relative timing constitutes an invariant parameter of a GMP, but that a different GMP is used to produce a walking versus running pattern because the relative timing characteristics were markedly different between the two gait patterns.

Researchers aligned with the dynamical approach to motor control (Diedrich & Warren, 1995, 1998) have also studied the same walk-to run transition sequence for the purpose of understanding

why people transition from a walking to running movement pattern as the task demands or environmental constraints change. Using a very similar experimental protocol, the authors observed a similar change in behavior but interpreted it in a very different way. In contrast to the explanation advanced by hierarchical advocates, the authors of these more recent studies interpreted their findings within the framework of the dynamical approach referred to as dynamic systems theory. They argued that transitions between movement patterns arise from the self-organizing properties of the human system and its desire to maintain stability. In order to achieve this stability, individuals adopt preferred patterns of coordination, or attractor states. In the face of changing task and/or environmental demands (e.g., increasing speed of the treadmill) that disrupt the stability of the preferred attractor state, individuals spontaneously transition from one attractor

state or preferred pattern of coordination to another attractor state that is more stable. According to dynamic systems theorists this transition does not require any conscious processing on the part of the individual or the need to switch to a different motor program.