It is widely held among cognitive aging researchers that older adults are at a disadvantage with respect to the division of attention between two or more concurrent tasks. Some researchers have attributed dual-task performance decrements to reduced processing speed with age while others have attributed declines in dual-task performance to the reduced efficiency of task coordination and control processes. Few researchers, however, have considered the possibility that age-related differences in dual-task performance may be related to underlying differences in the learning mechanisms supporting component task performance. Three studies were conducted which differed in the type of single-task training provided to young and old adult participants: Consistently mapped (CM), variably mapped (VM) and attenuated priority (AP) training. Skilled dual-task performance was then assessed as a function of both component task learning and age through a) the examination of initial and end-level skilled dual-task performance, b) transfer of learning to novel task combination and c) retention capability. It was predicted that type of component task training would moderate age-related differences in skilled dual-task performance. The results were confirmatory, however, were not completely consistent with initial predictions.