This article reviews the behavioral literature on the control of goal-directed aiming and presents a multiple-process model of limb control. The model builds on recent variants of Woodworth's (1899) two-component model of speed-accuracy relations in voluntary movement and incorporates ideas about dynamic online limb control based on prior expectations about the efferent and afferent consequences of a planned movement. The model considers the relationship between movement speed and accuracy, and how performers adjust their trial-to-trial aiming behavior to find a safe, but fast, zone for movement execution. The model also outlines how the energy and safety costs associated with different movement outcomes contribute to movement planning processes and the control of aiming trajectories. Our theoretical position highlights the importance of advance knowledge about the sensory information that will be available for online control and the need to develop a robust internal representation of expected sensory consequences. We outline how early practice contributes to optimizing strategic planning to avoid worst-case outcomes associated with inherent neural-motor variability. Our model considers the role of both motor development and motor learning in refining feed-forward and online control. The model reconciles procedural and representational accounts of the specificity-of-learning phenomenon. Finally, we examine the breakdown of perceptual-motor precision in several special populations (i.e., Down syndrome, Williams syndrome, autism spectrum disorder, normal aging) within the framework of a multiple-process approach to goal-directed aiming.

In this article, we examine the question of what information is processed during observational learning by evaluating a variety of methods, theories, and empirical data. Initially, we review work involving neuroimaging techniques and infant imitation. We then evaluate data from behavioural experiments involving adults, wherein a variety of attempts have been made to isolate the critical or minimal information constraining the acquisition of coordination. This body of research has included comparisons between video and point-light displays, manipulations to the amount and type of information presented in the display, the collection of point-of-gaze data, and manipulations to the task context in terms of outcome goals. We conclude that observational learning is governed by specific features of the model's action (i.e. motions of the end effector) and the task (i.e. outcome constraints) and, in contrast with traditional theoretical modelling, more global aspects of a model (i.e. the relative motions within and between joints) do not appear to be the primary method for constraining action execution.

Recently our group forwarded a model of speed-accuracy relations in goal-directed reaching. A fundamental feature of our multiple process model was the distinction between two types of online regulation: impulse control and limb-target control. Impulse control begins during the initial stages of the movement trajectory and involves a comparison of actual limb velocity and direction to an internal representation of expectations about the limb trajectory. Limb-target control involves discrete error-reduction based on the relative positions of the limb and the target late in the movement. Our model also considers the role of eye movements, practice, energy optimization and strategic behavior in limb control. Here, we review recent work conducted to test specific aspects of our model. As well, we consider research not fully incorporated into our earlier contribution. We conclude that a slightly modified and expanded version of our model, that includes crosstalk between the two forms of online regulation, does an excellent job of explaining speed, accuracy, and energy optimization in goal-directed reaching.

although most of the experiments reviewed here involved laboratory tasks such as rapid aiming and movement sequencing, the majority of the principles apply to motor control and learning in more complex situations. Thus, they should be considered when developing methods to train medical personnel to perform perceptual motor procedures with precision.

It has been proposed that, when learning a motor skill, individuals initially freeze degrees of freedom to simplify control. There is limited empirical evidence to support this proposition. We examined this issue by monitoring the performance of a non-skilled individual learning a soccer chip shot with his non-dominant leg over 9 days of practice (425 trials). Principal component analysis was used to examine dimensional change. The most dramatic change occurred at the hip, with the range of motion decreasing during the first 5 days of practice and then increasing thereafter. A reverse pattern was observed at the knee and ankle. While showing a progression in control from proximal to distal, a further phase was observed where primary control was passed back to the hip. The degree of linear coupling between the joints also increased with practice until day 5, after which independent control was observed. The number of controlled dimensions did not change across practice. Radial error decreased over practice and kinematics relating to the hip were most predictive of error, especially early in practice. Freezing degrees of freedom was a strategy implemented across the first half of practice, after which point-independent control was gradually restored enabling successful consistent performance.

No abstract

The acquisition of sensorimotor parameters that control goal-directed motor behaviors occurs by observing another person in the absence of efferent and afferent motor signals. This is observational practice. During such observation, biological motion properties associated with the observed person are coded into a representation that controls motor learning. Understanding the underlying processes, specifically associated with coding biological motion, has theoretical and practical significance. Here, we examined the following questions. Experiment 1: Are the underlying velocity characteristics associated with observed biological motion kinematics imitated? Experiment 2: Is attention involved in imitating biological motion kinematics? Experiment 3: Can selective attention modulate how biological motion kinematics are imitated/represented? To this end, participants practiced by observing a model performing a movement sequence that contained typical or atypical biological motion kinematics. The differences in kinematics were designed to dissociate the movement constraints of the task and the anatomical constraints of the observer. This way, we examined whether novel motor behaviors are acquired by adopting prototypical movements or coding biological motion. The kinematic analyses indicated the timing and spatial position of peak velocity were represented. Using a dual-task protocol, we attenuated the coding of biological motion kinematics (Experiment 2) and augmented coding using a selective attention protocol (Experiment 3). Findings indicated that velocity characteristics of biological motion kinematics are coded during observational practice, most likely through bottom-up sensorimotor processes. By modulating motion coding using 2 attentional protocols, we showed that bottom-up processes are influenced by input modulation, which is consistent with top-down control during observational practice.

An occlusion protocol was used to elucidate the respective roles of preprograming and online control during the quiet eye period of golf putting. Twenty-one novice golfers completed golf putts to 6-ft and 11-ft targets under full vision or with vision occluded on initiation of the backswing. Radial error (RE) was higher, and quiet eye was longer, when putting to the 11-ft versus 6-ft target, and in the occluded versus full vision condition. Quiet eye durations, as well as preprograming, online and dwell durations, were longer in low-RE compared to high-RE trials. The preprograming component of quiet eye was significantly longer in the occluded vision condition, whereas the online and dwell components were significantly longer in the full vision condition. These findings demonstrate an increase in preprograming when vision is occluded. However, this was not sufficient to overcome the need for online visual control during the quiet eye period. These findings suggest the quiet eye period is composed of preprograming and online control elements; however, online visual control of action is critical to performance.