This thesis set out to determine the optimum method of presenting multimedia instructional materials for learning a procedural task using a head-mounted display component of a wearable computer. There were four initial research questions; the first compared head-mounted display technology to traditional methods for following procedural task instructions. The second compared subtypes of head-mounted display for following task instructions and a third investigated possible differences between using an opaque and see-through head-mounted display for the task. A fourth research question was to involve exploratory experiments into the optimum configuration of text and vocal instructions on a video demonstration of the type used on a headmounted display. The research programme comprised of two phases, the first phase investigated the first three questions. Results from the first phase indicated no significant difference in performing a procedural task on a head-mounted display in the conditions that investigated first and third research questions. Due to technological differences between the head-mounted displays the second research question could not be addressed since comparisons between these displays were not feasible. This inability to make comparisons between subtypes resulted in second phase of experiments concentrating on the fourth research question; the investigation of the optimum presentation of multimedia learning materials on a video demonstration. The research in the second phase tested whether present guidelines for demonstrations for learning factual knowledge were applicable for creating multimedia demonstrations for learning procedural tasks. Results from the second phase of experiments challenge the use of the above guidelines for producing video demonstrations for learning a procedural task.