Comparing pointing performance of mouse and eye-gaze input system
Metadata[+] Show full item record
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] The purpose of this study is to compare the pointing performance of mouse and eye-gaze input system to find the characteristics of the eye-gaze input system. By using a human-in-the-loop simulation, we could collect the participants' reaction time, accuracy and cognitive workload to evaluate the participants' ability to detect abnormal events during a visual searching task between different input systems. In this experiment, subjects were arbitrarily divided into two groups labeled mouse input system and eye-gaze input system. Fitts' law model, signal detection theory (SDT) and NASA task load index (NASA-TLX) were used to assess and compare participants' performance between different input systems. First, based on the Fitts' law model, we wanted to verify that whether there is a specific relationship between reaction time and index of difficulty (ID) for both mouse and eye-gaze input systems. The results presented that ID influenced the reaction time of mouse input, but the eye-gaze input system is less influenced by ID. For the eye-gaze input system, there was not linear pattern between reaction time and ID. Although the increase was small, reaction time had a certain degree of increase compared to different ID. On the other hand, the reaction time difference between mouse and eye-gaze became larger and larger as ID became higher. So there was a significant difference in reaction time between mouse and eye-gaze input systems. The eye-gaze input system had more distant click advantages than the mouse input system on the large screen and virtual reality environment. Second, for the mouse input system, normally if we reduced reaction time, the accuracy would decrease. However, the eye-gaze input system maintained the same accuracy but had a faster pointing time for operators. The mouse input system has more miss response, but eye-gaze input system has more false alarm response. So it is recommended to use either input system in different environments. Meanwhile, we should decrease the false alarm ratio to improve the accuracy of the eye-gaze input system. Third, people typically think the mouse input cognitive workload will be higher instead of eye-gaze. However, eye-gaze input maintained the same cognitive workload compared with mouse input system. All the findings can be utilized to provide deeper insight into user behavior analysis regarding different input systems. The results showed the potential of eye-gaze input as a valuable tool for advanced human-computer interfaces if participants are well-trained to use eye-gaze input system in which the eye tracking device accuracy is good enough.