Related Experiment Video
Updated: Jun 2, 2026
09:27
Using Eye Movements Recorded in the Visual World Paradigm to Explore the Online Processing of Spoken Language
Published on: October 13, 2018
Mode-of-disparities error correction of eye-tracking data
Yunfeng Zhang1, Anthony J Hornof
1Department of Computer and Information Science, University of Oregon, 1202 University of Oregon, Eugene, Oregon 97403-1202, USA. zywind@cs.uoregon.edu
Behavior Research Methods
|April 14, 2011
Summary
Eye tracking often has errors. This study introduces a new method to correct systematic gaze errors, improving fixation accuracy even with irregular object arrangements.
Area of Science:
- Cognitive Science
- Human-Computer Interaction
- Psychology
Background:
- Eye-tracking technology is crucial for understanding visual attention.
- A common challenge is the disparity between recorded and actual gaze locations, known as systematic error.
- Existing methods may struggle with complex or non-uniform experimental setups.
Purpose of the Study:
- To present a novel method for reducing systematic error in eye-tracking data.
- To accurately restore fixation points to their true locations.
- To demonstrate the method's reliability, particularly in irregularly arranged visual environments.
Main Methods:
- Calculating disparities between recorded fixations and nearest visual objects.
- Employing the annealed mean shift algorithm to identify the mode of these disparities.
- Utilizing the mode to determine and correct the systematic error's magnitude and direction.
Main Results:
- The proposed method effectively reduces systematic error in eye-tracking data.
- Fixations are reliably restored to their true locations.
- The method demonstrates efficacy even when visual objects are not arranged in a regular grid.
Conclusions:
- The annealed mean shift-based approach offers a robust solution for eye-tracking systematic error correction.
- This technique enhances the precision of eye-tracking data, especially in complex experimental designs.
- Accurate gaze data is vital for reliable insights in various research fields.
