Binocular Coordination in Reading When Changing Background Brightness

Abstract

INTRODUCTIONWhen we look, we fixate on an object of interest with both eyes. The process of aligning both focus points is referred to as binocular coordination, the result of this process is termed vergence. That is, a vergence point is defined as the position where both lines of sight meet. Understanding the alignment of both lines of sight is important also for the understanding of binocular vision, that is, the motor coordination of the eyes as well as the perceptual coordination like, for example, in image fusion. In visual tasks, vergence points are expected to occur where an image is depicted. Therefore, when presenting images on a screen, the fixation location of both eyes is measured on the screen. The resulting difference between fixation locations of both eyes is referred to as fixation disparity. That is, fixation disparity is the difference between the two lines of sight at a given distance, which informs about the spatial accuracy of aligning both lines of sight.Binocular coordination is performed also in seemingly two-dimensional tasks like, for example, in reading. This raises the question of whether two eyes are better than one, as Radach and Heller (1999) put it. With regard to fixation disparity in reading, there are contradictory results. Liversedge et al. (2006) reported that about half of the fixations accompanying reading are unaligned. The authors characterised these fixations as crossed or uncrossed, depending on their disparity: fixations were counted as crossed when the right eye fixated more than one character left of the left eye, and as uncrossed when the right eye fixated more than one character right of the left eye. Expressed as vergence, crossed fixations were described as a vergence point spatially in front of the screen, and uncrossed fixations as a vergence point behind the screen. Liversedge et al. (2006) found more uncrossed fixation disparities than crossed ones, that is, a mean vergence point behind the screen.In contrast, a dominance of crossed fixations and hence, vergence points, which are typically in front of the screen, was reported by Nuthmann and Kliegl (2009). With regard to binocular coordination, one might wonder how image fusion can take place on the basis of such huge amounts of not aligned fixations. In order to understand binocular visual functioning, it is therefore important to distinguish systematic variations of fixation alignments from random variability.There were already various reasons suggested as potentially being responsible for the differences in the findings of Nuthmann and Kliegl (2009) as compared to : differing eye-trackers, calibration procedures, viewing distances and also letter sizes. The former used a chin rest, the latter a bite bar, and participants had to read in different languages (German vs. English). These differences and several more (e.g., glasses, contact lenses, eye colour, PROCEEDINGS OF THE LATVIAN ACADEMY OF SCIENCES. Section B, Vol. 71 (2017)

Do pupil-based binocular video eye trackers reliably measure vergence?

“…The vergence angle and the distance to the binocular fixation point are used to investigate depth perception and binocular coordination in vision science (Enright et al, 1987a(Enright et al, , 1987bWagner, Ehrenstein, & Papathomas, 2009;Wismeijer, van Ee, & Erkelens, 2008;Wismeijer & Erkelens, 2009) and binocular coordination in reading (Kirkby, Blythe, Benson, & Liversedge, 2010;Köpsel & Huckauf, 2017;Huckauf, 2018;Nuthmann & Kliegl, 2009). There is a debate about fixation disparity in the literature on binocular control in reading.…”

“…The stimuli in Kirkby's lab (white on a black backgound) differed from the stimuli (black on a white background) used in the labs that reported the unaligned crossed fixations with a pupil-based video eye tracker (Jainta et al, 2010;Nuthmann & Kliegl, 2009), Vernet and Kapoula (2009) did not provide this information. Köpsel and Huckauf (2017) used dark and light backgrounds and were able to replicate the results of both Nuthmann and Kliegl (2009) and Liversedge et al (2006) with a pupil-based video eye tracker (SMI iView X hi-speed 1250). Huckauf (2018) added combinations of calibrations with dark and light backgrounds with dark and light stimuli and report that when the calibration background matches the one during reading, fixation disparity was close to zero deg.…”