The motor control of 49 unmedicated boys clinically diagnosed with ADHD, case-matched with 49 non-ADHD boys, was assessed while playing Crash Bandicoot I, a Sony(TM) Playstation platform computer video game. In Crash Bandicoot participants control the movements of a small-animated figure through a hazardous jungle environment. Operationally defined measures of motor control were designated by (1) the stage of the game completed (ie, the number of obstacles successfully passed) before losing the figure's 'life', (2) the level of complexity that the stage represented and (3) the time taken to get to that point during the video game play. These measures were assessed under contrasting conditions of low or high working memory and distracter loads. Four tasks were administered (totalling 12 trials), incorporating both with and without distracter conditions. For those trials with the distracter, a segment of the television show The Simpsons was simultaneously played on a television screen adjacent to the computer game monitor. A 5-way MANOVA revealed that ADHD boys took less time to complete their trials under the direct condition (ie, no working memory load) on Crash Bandicoot, compared to their matched non-ADHD peers. When the task required additional working memory, however, the ADHD boys took significantly longer. Cumulative frequency plots of game performance revealed that in terms of the number of obstacles completed, the control participants successfully navigated more obstacles on the low working memory load task than the ADHD participants, but that the performance of the two groups was less distinguishable on the high working memory load task. The findings have implications for assessment and management of children with ADHD.