An eye movement study for sustained tracking of objects
This thesis is a study of the eye movement mechanisms involved in the sustained pursuit of an object. This is done by examining the eye movements involved in locating an object, and subsequently tracking a moving target. Experiments with a range of subjects to collect data on their eye movement have been performed. Based on the results of these data, statistical analysis have been performed to identify significant parameters. ^ Various methods have been developed in this thesis with these goals in mind. A custom program was coded to facilitate the experiments and obtain the results in an easily accessible way. The information coming from the eye tracker computer has been processed through the program in order to obtain the position of the eye as the raw data for the experiment. ^ This program integrates data acquisition and image display functions, so the output is a simple correlation of where the eye is looking, and where the object being tracked is. The latency time, which is the delay between the presentation of the object and the time that the eye moves to is used as the primary dependent variable. ^ Having obtained the raw data, parameters such as the duration of the experiment, speed and direction of the object while being displayed, the duration and the distance and relative position between the point where the object disappears and where it appears again were studied statistically to examine which of these have an influence on the latency time. ^ From the raw data, a matrix has been created with the average values of the latency times for each subject and for each case having. The values that statistically exceed the averages at a 95% and 99.5% confidence level were highlighted, and were used for developing conclusions. ^ Further more, once that it was observed that certain movements results in higher or lower latency times, a new method for analysis was developed to segment the experimental cases depending on the combination of the previous movement of the target and the next one. The latency times have been classified by the program in one of the eight different movement combinations for each of the subjects, and statistical analysis have been performed to highlight the values that are significant. ^ In conclusion, the independent factors which are (1) duration of the experiment, (2) time displayed, (3) speed and (4) direction of the object displayed, (5) time that the object is hidden, (6) position and (7) distance of the relative position between the point where the object disappears and where it appears again do not have a general correlation with latency time. However, the range of responses is very broad and some subjects appear to have responses that are much faster or slower than average values. It is interesting to note that this was not manifest across all scenarios. Further testing with a larger pool of subjects is recommended to refine these conclusions.^
Engineering, Computer|Psychology, Experimental
Jesus Garcia Galvez,
"An eye movement study for sustained tracking of objects"
Dissertations and Master's Theses (Campus Access).