Hello all,
Currently I am working with eye-movement data collected using the Tobii Pro Glasses 3. We have a clinical group and a healthy control group who performed a series of tasks while equipped with these eye-tracking glasses. Task durations were not controlled as this was also a measure of interest, so participants took different amounts of time (seconds) to complete the tasks (in general, the clinical group taking a longer time to complete the tasks than the healthy control group). We want to compare the number of saccades and fixations performed for each task between the two groups. Since the time taken to complete the task would naturally contribute to the number of eye movements performed, is there a way to perform this between-group comparison while also taking in account this variation in time? Any suggestions would be very appreciated!
Thanks,
Saee.
Hello,
To account for task completion time differences, normalise the eye movement data by calculating the rate of saccades and fixations per second for each participant. This allows for a fair comparison between the clinical and healthy control groups.
Hope this helps.
Number of fixations (not saccades) and fixation duration (first fixation duration, single fixation duration, gaze duration, total fixation duration) may be computed on single (target) word basis. Word skipping and regressions may be computed too. For region based measurements (e.g., sentences), consider first pass, second pass, and total reading time. Task duration (longer for the clinical group) depends both on the number of fixations and their duration. That's the difference between the groups (i.e., your main result, not a bias to be controlled).
Try also this: Article The Thirty Fifth Sir Frederick Bartlett Lecture: Eye movemen...
- Badges
- Science topic