There has been a conspicuous absence of research directed to the question of whether older adults have the capability to improve their dual-task performance through practice and training, and whether such improvements might parallel those exhibited for younger adults or instead narrow or eliminate the age-related gap in dual-task performance. It is this question which forms the basis of our research program on the influence of training strategies on the learning, transfer and retention of dual-task processing skills for young and elderly adults. The major aims of our research program include the following; (1) To examine the hypothesis that training strategies, such as variable priority training (Gopher, in press), which emphasize the flexible coordination among two or more tasks in multi-task settings, will enhance the rate of learning and level of mastery exhibited on trained tasks as well as the retention of performance over extended periods of time. Furthermore, we predict that such training strategies are uniquely suited for older adults given their demonstrated difficulties in metacognitive skills such as self-monitoring and information. Thus, we predict a reduction in age-related dual-task performance differences with the use of variable priority training strategies. (2) To examine the hypothesis that dual-task training, particularly with variable priority training strategies, will lead to the improvement in a number of aspects of processing including the automatization of the trained tasks, the improvement in the efficiency of dual-task processing strategies such as time-sharing and time-switching, and improvements in the efficiency of elementary processes relevant to dual-task performance such as perceptual speed, working memory, and various attentional processes. (3) To examine the hypothesis that the scheduling of performance feedback, in the form of adaptively fading feedback with practice, will enhance the transfer and retention of dual-task processing skills and reduce the age- related decrement in dual-task performance. To test this hypothesis we will adapt some of the feedback scheduling techniques that have been successfully used in the motor and verbal learning literatures to the arena of dual-task processing.
