Computer Science and Informatics
Permanent URI for this community
Browse
Browsing Computer Science and Informatics by Author "Fouche, Rouxan Colin"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
Item Open Access The usability of natural user interfaces for gameplay(University of the Free State, 2015) Fouche, Rouxan Colin; Beelders, T. R.; De Wet, L.English: This study aimed to determine to what extent the usability of a two dimensional game was influenced by the use of a Natural User Interfaces (NUI) as opposed to a traditional keyboard and mouse combination. Two multimodal NUIs were investigated during the study. The first NUI combination (BCIG) made use of the Peregrine gaming glove for the activation of commands, combined with the Emotiv’s accelerometer for control of the cursor. The second NUI combination (BCIF) made use of facial expression recognition, offered by the Emotiv Brain Computer Interface (BCI), as a method of command activation in combination with the Emotiv’s built-in accelerometer for cursor control. A shooting genre game was developed and three tasks were included during development to simulate gaming actions. The first task used only stationary targets, the second task used predictable moving targets, whereas the third task made use of unpredictable moving targets. Since the Emotiv BCI allows for customisation of the accelerometer sensitivity settings, a pilot study was conducted to determine whether the low, medium or high sensitivity setting would provide the best cursor control. The low sensitivity resulted in the fastest gameplay overall as well as the least number of errors. It could thus be concluded that the lowest setting is the optimal setting since it provided the most efficient control for three out of the four metrics tested. After implementing this result, user testing, which involved 5 sessions per participant (n=18), was conducted. Data for three metrics was gathered during user testing, which included data on effectiveness, efficiency and learnability. Post-test questionnaires were administered to assess the level of user satisfaction with each NUI. The results of this study indicated that there is a difference between the usability of the traditional input combination, the keyboard and mouse, and the two NUIs investigated in this study. With regard to the effectiveness and efficiency metrics the traditional input combination was found to be the best, closely followed by BCIG. The three interfaces showed dissimilar levels of improvement, with the traditional keyboard and mouse combination showing the least, followed by BCIG, while the best improvement was noticed for BCIF. By analysing the subjective data gathered from post-test questionnaires, it was found that participants initially enjoyed using BCIG, and after several sessions their level of satisfaction improved. In comparison, the participants initially experienced a slightly negative feeling towards BCIF, which then improved over several sessions to a positive overall response. In conclusion, the keyboard and mouse combination provided far more effective and efficient input, with one exception being the fast command activation when making use of the Peregrine glove, where the two interfaces compared well. It was found that a significant obstacle in the way of NUIs is the existing skill and acceptance that computer users have with the traditional interface combination. Consequently, for individuals to accept and migrate to a more natural interface the new interface will have to provide more effective and efficient input than what is already achievable with the keyboard and mouse combination.