Redefining the developmental stages of overarm throwing action from a dynamical system theory perspective

Considerable research has focused on identifying the series of stable states in technique that occur when practising complex motor tasks. From an applied perspective, forming a model of motor learning in this way informs interventions to elicit the most efficient and effective practice. From a theoretical perspective, models of motor learning provide an insight into qualitative and quantitative states that reflect how movement is controlled. Assimilating these two perspectives, the overall aim of this research was to increase understanding of the key characteristics of technique change during learning a complex whole-body movement skill. In order to achieve the overall aim, two experimental studies were conducted. The first was a longitudinal learning study in adult participants. Non-dominant overarm throwing action was practiced over a 3-week period consisting of 9 sessions. This was followed up with a cross sectional study in which technique in dominant overarm throws at 6, 10 and 14 years of age was examined. The application of different approaches to analyse technique changes provided a platform to better understand motor learning. Findings showed that the use of a macroscopic collective variable was able to provide an overarching view of dynamical changes of the system with practice and age. Alternatively, changes in joint range of motion was found to be individual specific among adults and children with no clear direction of change related to practice or age. From a practical perspective the components model (Roberton & Halverson, 1984), particularly the step action, underpinned the macroscopic changes and provided a tool that can be more easily applied by practitioners and educators to facilitate and fast track learning. The resulting research philosophy emerged: the application of a theoretically underpinned, multi-disciplinary approach to quantify technique change during learning has the potential to move researchers closer to a generalised model of motor learning