Training Content Variability and the Effectiveness of Learning: An Adult Age Assessment

Making training easy and rapid has been a goal for those who train older learners. Even though they may initially produce easy and rapid performance change, however, some training conditions may actually hinder learning by reducing subsequent retention and generalization (Schmidt & Bjork, 1992). To assess this, older and younger adults were trained on an algorithm for mentally squaring two-digit numbers. Half of our participants received training in a high-variability content condition involving problems widely dispersed across the range of decade and 1's digits of two-digit numbers; the other participants were in the low-variability content condition that involved training problems with a narrow range. For both age groups, high-variability training resulted in inferior performance at the end of training, compared to low-variability training. Consistent with Schmidt and Bjork, however, high-variability trained younger adults were marginally better at retention, and were significantly better on nontrained transfer problems. In contrast, high-variability trained older adults did not differ from their low-variability trained agemates at either retention or generalization.