Comparing the efficiency of gaze as input modality with mouse

In the modern era, computers are an integral part of our lives. Few decades back, when computers were novelty, we used to communicate with computers with teletype interfaces and it used to be a specialist's job to do so. With the advancement of computers, there has been many advancements in the way of how we communicate with the computer. Billions of people have access to personal computers today. Most interact primarily with hand based external devices like keyboard and mouse. With computers becoming a part of our daily lives, new interfaces of communication such as gesture-based interface, speech-based interface, virtual reality etc. are being researched. Gaze is such a novel input modality. In this thesis we are comparing the efficiency of gaze input modality with a traditional input like mouse. This comparison is made in the context of online gaming of collaborative nature. While trying to compare and analyse gaze with mouse, Space Explorer game was developed. The storyline is that the two players are exploring the space in a spaceship, but they have to shoot asteroids to make way for them. The target setter identifies and sets the target asteroid. The said asteroid is highlighted in both player's screens. There were fourteen participants from the Tampere University, who had opted in to participate in the experiment. They were assigned a role of being target setter and target shooter randomly. Target setter uses gaze to set the target and target shooter uses mouse to shoot the target. After that roles are switched. Then the experiment is conducted with both target setter and target shooter doing the same tasks with mouse. Then the participants switched roles so that all participants experienced both the roles with mouse in- put. At the end of the experiment, the participants filled up a questionnaire which focused on their user experience. This thesis uses research methods of interactive design, online collaborative gaming, software engineering, prototyping and user study. The role of being target setter was evaluated. The target setting role was performed by all the participants, both with mouse and gaze input. It was found that the participants took significantly more time to complete the task with gaze than with mouse. Participants on an average took 3.4 seconds to do the task with gaze as opposed to 1.7 seconds with mouse. Subjective experiences show that the preferred input was mouse. Participants said that the familiarity of mouse, along with the accuracy and smoothness made mouse the clear winner. Participants also pointed out many other shortcomings with gaze technology. They said that the tracking was jittery and error prone. A rather common feedback was that the participants sometime had to look away from the target to correct the error offset caused by the tracker, making it frustrating. Few participants though said that their preferred method of interaction was gaze and they liked the novelty and interestingness of the technology. Overall mouse emerged as a clear winner