Age effects on the transfer of sequence knowledge between different types of movement

The goal of the current study was to determine whether age influences the ability to transfer sequential knowledge between two different types of motor responses. During practice of a discrete sequence production task, participants responded either by unimanual key presses (KP) on a standard computer keyboard, or by moving a lever with a flexion extension (FE) motion of their right arm. Sequence knowledge was then tested with the other type of responses. Stimulus presentation was identical over the whole task. Sequence learning theory suggests that performers first develop a cognitive representation and with additional practice a motor representation. Based on this insight and on previous research, we hypothesized that elderly would be able to transfer sequence knowledge between the two response modes. Because KP responses are less precise, we expected this condition to be more cognitively controlled than the more difficult FE condition, and therefore also hypothesized more transfer from practice using KP followed by a retention test using FE than vice versa. We tested 32 right-handed elderly (65 - 74) and 27 young people (18 - 30). Individual characteristics were described using the MOCA, a visuospatial working memory task, a digit symbol substitution task and a general health questionnaire. The experiment started with familiarization with KP and FE. Then, two 6-element sequences were practiced with KP or FE for a total of 288 trials. Participants were instructed that two fixed sequences were presented, but not that a test on transfer to another response mode would follow. After an explicit sequence knowledge questionnaire, the test phase with the remaining type of movements followed. The test phase consisted of one block of 24 familiar trials and one block of 24 random trials. Transfer was defined as the percentage speed difference between the familiar and random test blocks. After FE practice, both age groups showed transfer of sequence knowledge. This effect was larger for young participants than for elderly. After KP practice, only young participants showed signs of transfer, which was smaller than the amount of transfer after FE practice. Because there was no difference in the amount of explicit sequence knowledge between the practice conditions, we conclude that there is either more implicit spatial sequence learning in the FE than in the KP response mode, or that KP responses are a more sensitive measure of implicit sequence knowledge. We found that elderly are indeed able to transfer sequence knowledge but to a lesser extend than younger people and that in both age groups response mode plays an important role in measuring such effects