|
Research [Home | Lab Members | Research | Publications | Courses | Pictures] |
|
Topics of Interest:
The Influence of Movement Dynamics on Motor Planning
The Influence of Movement Dynamics on Motor Planning
[back] When we interact with our environment, there are an infinite number of movement paths to accomplish any task. Despite this redundancy, we often make repeatable movements; our neuronal circuitry, skeletal biomechanics and musculature work together in a concerted effort to achieve this stereotypy. Typically, motor planning is studied for a two degree of freedom arm performing a horizontal planar task. Little has been done to extend our understanding of motor planning to include a gravitational force or complex dynamics. We are interested in better understanding the origins of movement stereotypy. In particular, we are interested in the roles of gravity and movement dynamics in the planning and execution of arm movements. The goal of our research is to extend the knowledge of motor planning and behavior from the study of planar movement to more complex, three dimensional movements.
Motor Adaptation to Transient Perturbations of Reaching Movements
[back] Motor theorists, psychophysicists, and neurophysiologists have several notions of how sensed error induces learning, even though actual quantification of specific trial-by-trial learning has been scarce. Hypothesized training signals include motor error, sensory-error driven feedback control (, and sensory error itself. Desired signals supervise these rules; learning may instead take place via unsupervised critics of movement appropriateness, as in reinforcement learning. To more precisely identify the causal relation between specific errors and incremental adaptation, we perturb subjects with short pulses of force (introduced 2-7 cm into movement, 6-18N in strength) as they make reaching movements while holding the handle of a robotic manipulandum. Our results show all pulses yielded similar changes in predictive control regardless of the position at which the pulse was experienced. If forces pushed to the left, people reliably adapted by moving to the right; if forces pushed right, people reliably adapted by moving to the left. Furthermore, subjects adapted their predictive control starting very early in the next movement. This suggests that the adaptation of predictive control cannot depend narrowly upon the position at which unexpected feedback occurs. While the kinematic error induced by the forces sensibly scaled with the magnitude of the perturbation, the adaptive response was insensitive to the pulse magnitude, whether perpendicular displacement was measured early, middle, or late in movement. Neural models that broadly encoded movement velocity and adapted in proportion to motor error mimicked human insensitivity to pulse location. Current supervised and unsupervised theories of motor learning, however, all presume that adaptation is proportional to error and therefore fail to mimic human insensitivity to pulse magnitude. We conclude that single trial adaptation to force pulses reveals a novel adaptive strategy that aims to coarsely counter the sign, rather than the magnitude, of movement error. Work is currently underway to determine the sensory
experiences necessary to induce a transition from adaptation proportional
to error (as in Scheidt et al., 2001) to the coarse, insensitive
adaptation discovered here.
Effects of Generalization when Learning
Complex Dynamic Environments
[back] People routinely learn how
to manipulate new tools or make new movements. This learning requires the
transformation of sensed movement error into updates of predictive neural
control. Here we demonstrate that the richness of motor training
determines not only what we learn but how we learn. Human subjects made
reaching movements while holding a robotic arm whose perturbing forces
changed directions at the same rate, twice as fast, or four times as fast
as the direction of movement, therefore exposing subjects to environments
of increasing complexity across movement space. Subjects learned all
three environments and learned the low and medium complexity environments
equally well. We found that subjects lessened their movement-by-movement
adaptation and narrowed the spatial extent of generalization to match the
environmental complexity. This result demonstrated that people can
rapidly reshape the transformation of sense into motor prediction to best
learn a new movement task. We then modeled this adaptation using a neural
network and found that, to mimic human behavior, the modeled neuronal
tuning of movement space needed to narrow and reduce gain with increased
environmental complexity. Prominent theories of neural computation have
hypothesized that neuronal tuning of space, which determines
generalization, should remained fixed during learning so that a
combination of neuronal outputs can underlie adaptation simply and
flexibly. Here we challenge those theories with evidence that the
neuronal tuning of movement space changed within minutes of training.
The effect of attention on the feedback and adaptive control of arm
movement dynamics. [back] To understand the normal goal-directed motor behavior of healthy individuals, it is important to allow subjects to utilize freely any available cognitive resources to accomplish the motor task. When learning a novel motor task in a natural environment, such as learning to play the piano, humans can allocate as much attentional focus to the task as they desire. Most motor control and adaptation studies examine and control only the motor aspects of the task, other variables, such as attention, are less controlled. This allows the functional contribution of specific motor areas to the control of movements and acquisition of motor skills to be well studied, however, the contribution of higher-level cognitive areas cannot be characterized. It is currently unknown as to how attention influences the trial-by-trial learning of a novel motor skill. One approach to interrogating important contributions of multiple areas to a particular task is to overload the system with too much information or processing demand by having subjects perform two tasks simultaneously. Previous dual-task studies have revealed a decrement in performance in novel motor tasks when human subjects are distracted with a secondary task, but no decrement in performance in practiced motor tasks. Recent fMRI data have shown neural anatomical correlates of activation and interaction between executive control/attentional areas and motor areas during the initial stages of a motor learning task. However, it is unknown as to what the specific functional contribution of attention is on the online control of the current movement (feedback control) and how errors in the current movement inform future movements (predictive control). We are currently performing a series of experiments that characterize how dual-task interference influences the control of the present movement, the change in predictive control of the next movement, and of the eventual formation of long-term motor memory. Taylor, J.A. & Thoroughman, K.A. |