Instruction processing in young and older adults: Contributions of memory span

Background: Age-related changes in cognition and in particular, working memory, can impact older adults' abilities to comprehend linguistic information. Many investigators have undertaken the study of age effects on language comprehension, but confounding variables, such as vocabulary level, general knowledge, and episodic memory ability limit what can be inferred about linguistic processing in ageing. Aims: The purpose of this study was to investigate young and older adults' performance on a language-processing task that assessed ability to follow instructions. Despite having ecological validity, little attention has been paid to age effects on the processing of procedural information. The use of the instruction task allowed for an investigation of age effects on language comprehension while mitigating the effects of thematic knowledge, vocabulary and episodic memory. Methods & Procedures: A total of 37 older adults (M = 72.1 years) and 41 young adults (M = 22.5 years) received three measures of verbal memory: digit span, word span, and listening span. In addition, they were administered an experimental instruction task requiring participants to sort coloured pills into pill containers in response to spoken instructions. Information load of the instructions was manipulated by varying the number of actions per instruction and the number of components to be remembered per action. For example, the instruction Take three pills on Monday and two on Tuesday has two actions, and each action has two components to remember (number of pills and day). The dependent variable was participants' performance accuracy in following the instructions. Outcomes & Results: Significant age effects were observed on the experimental instruction task, as well as on word and listening span measures. As the information load of the instructions increased, accuracy decreased for both groups, although this effect was greater for the older adults. When comparing instructions that had the same number of total components to be remembered, but differed in how these components were structured, participants performed more accurately when the instruction contained fewer actions, even if each action had more components to remember. Digit span was a significant predictor of performance on the instruction task, together with age accounting for more than half of the variance. Conclusions: The results of this study demonstrate the expected age effects on working memory span capacity, and illustrate the effect of span capacity on following verbal instructions. From a practical perspective, these findings suggest that when a procedural instruction loaded with content is presented to an older adult, processing will be enhanced when it requires fewer actions.