Eye Movement Parameters as Mediators of Virtual Environment and Visual Symptoms: A Structural Equation Modeling Approach

Eye movement parameters have been widely utilized as parameters of visual symptoms while immersing in the virtual environment. The purpose of this study was to identify several eye movement parameters that can predict the visual symptoms by utilizing a structural equation modeling approach (SEM). 6 males and 6 females were asked to perform a multidirectional Fitts’ Law task in 3D TV virtual environment. Several eye movement parameters such as number of fixation, pupil size, accuracy, saccade duration, revisited fixation duration, time to first fixation, and eye movement time were constructed as under one mediator latent variable named eye movement parameters. In addition, this latent variable was placed between virtual environment, index of difficulty, and visual symptoms. SEM showed that number of fixation, saccade duration, and eye movement time were three significant predictors for visual symptoms under the virtual environment. These findings can be utilized by human factors engineers to predict the visual symptoms in other virtual environment.