A method of creating and/or modifying a video for use in a head-mounted display while a user of the head-mounted display is carrying out a task that is illustrated by the video. The method may include processing an initial video to automatically determine when the initial video does not cover key aspects of the task and thus when video mashing is desired. When video mashing is desired, a supporting video is obtained to cover the missing key aspects, and the initial video and the supporting video are automatically mashed, result in a composite video for subsequent display on the head-mounted display.