We present preliminary results of an experiment to compare gestural, touch, and mouse interaction using Fitts' law. A total of 15 participants were asked to select 100 targets as quickly and accurately as possible using each technique. Selection of targets was split into rounds of 20 (separated by a short break) in order to evaluate whether fatigue affected performance or whether performance improved/declined over time. The results found that gestural interaction performed much worse than touch and mouse interaction and recorded 3 times as many missselections. The poor results for gestural interaction were attributed to participant unfamiliarity and inaccuracies of the gesture-sensing device (Microsoft Kinect). Touch interaction performed comparably with mouse interaction although suffered with smaller targets due to occlusion and the impreciseness of a finger compared to a mouse cursor. Overall, performance remained fairly consistent over subsequent rounds. Fatigue did not have any effect.
|Number of pages||4|
|Publication status||Published - 1 Jan 2013|
|Event||25th Australian Computer-Human Interaction Conference - |
Duration: 25 Nov 2013 → …
|Conference||25th Australian Computer-Human Interaction Conference|
|Period||25/11/13 → …|
- Fitts' law