Digital Manufacturing technologies can yield geometrically accurate dynamic assembly sequences to be used as work instructions. An independent groups experiment was carried out in order to investigate the effects of different instructional media on performance on a small scale mechanical assembly task. Twenty four participants completed the assembly task a total of five times on consecutive weekdays. Three types of unimodal instruction sets were designed and delivered via a laptop computer — text only; static CAD diagrams and CAD animation. Build times were recorded for each participant and plotted as a learning curve. Results suggested that the use of animated instructions can reduce initial build times, as the mean build time at build one was 37% and 16% quicker than the text and diagrams groups respectively. The beneficial effect diminished after the first build, however, the graphics (diagrams and animation) groups continued to yield quicker mean build times up until build 3. Results are discussed in light of cognitive theories relating to how we process instructional information.