|
|
  |
|
| |||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||
|
Time Scales and Variability in Motor
Learning and Development |
 |
||||||||||||||||||||||||||||||||
|
In recent papers we have developed a
dynamical framework of motor learning and development in which the
time scale of change is central to revealing the transitory and
persistent processes of adaptation and learning, respectively. This
perspective is derived from the principles of dynamical systems that
growth-decay at a constant rate will determine the time scale of
change close to fixed points. It has been shown that these
assumptions provide a principled framework to decompose the
performance dynamics of adaptation, learning and development. The changes in attractor dynamic of learning and development also lead to changes in the pattern of intra-subject variability from repeated attempts to perform the same task. Traditionally, the variability of motor output has been interpreted as noise but an increasing number of experiments has shown that white noise is rarely if ever the structure of motor output variability. The potential of system filters and adaptive processes also make it difficult to infer the level of the noise in the sensori-motor system from the level of noise in the motor output. |
There are, however, many different measures of movement variability. One discerning characteristic of these measures is that variability can be calculated over different time scales that include within-trial, between-trial of the same practice session, across sessions of days, weeks, months and years. These different time scales of variability may reflect different processes of learning and adaptation over the lifespan as they can for the mean change in performance over time. The within-trial variability of continuous tasks, such as maintenance of posture, reflect the moment to moment time scale of the variability of the attractor dynamics and has been shown to be highly sensitive to a range of individual subject and task condition variables. The estimate of within-subject variability over trials of different time frames gives emphasis also to the initial conditions of attractor formation and has been shown to be sensitive to the health and functional status of individuals. In both cases it seems useful to distinguish the stability of the task dynamics from the noise even though both contribute to the variability of output. In this paper we explore the theoretical and practical implications of the role of time scales and variability in motor learning and development. | ||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||||
