Critical Variables in Adaptive Motor Skills Training

Abstract

Automatic adaptive training procedures for motor skills learning have been suggested for several years. Adaptive training is a closed-loop system in which some aspect of the student's performance is monitored and used in a computer algorithm to adjust the difficulty of the training task. Only limited research exists on the evaluation of these procedures both in basic laboratory motor learning tasks and in applications to flying training using synthetic flight trainers. In this paper critical variables dealing with the performance measurement procedure, the choice of the adaptive variable, and the appropriate adaptive logic are evaluated. The results of a series of laboratory studies are reviewed, and several suggestions for additional research as well as a general model of instruction are discussed. It was concluded that the most effective automated adaptive motor skills training probably should use multivariate performance measurement schemes which manipulate stimulus-related adaptive variables in connection with a closed-loop adaptive logic model that is optimized for individual differences.