Abstract
Purpose :
Contrast sensitivity is a highly useful metric of visual ability, but has classically been difficult, time-consuming, or impossible to assess in individuals with cognitive impairment. We tested whether a rapid gaze-based task can be used to assess contrast sensitivity in both healthy participants and non-verbal children with brain injury.
Methods :
We first measured contrast sensitivity in sixty neurologically healthy participants with an eye-tracking-based task. This task displays drifting band-filtered noise targets and determines when the participant is smoothly tracking a target with their eyes, then adjusts the target’s spatial frequency and contrast in real time until it is no longer visible. Fifteen thresholds were measured in order to form a high-resolution contrast sensitivity function (CSF). We then measured CSFs in three children, two of whom had brain injury. For these participants, the task was modified to use scrolling full-screen noise, and the tester could manually toggle a cartoon landscape to regain their attention (a form of “bait and switch”).
Results :
Figure 1 depicts all corrected (blue) and uncorrected (red) CSFs measured from participants with refractive correction, ordered by uncorrected LogMAR eye chart acuity from best (lightest) to worse (darkest). These acuity scores are also given in red in the bottom-left corner of each plot. Uncorrected LogMAR acuity was strongly correlated with mean threshold radius (r = -0.89), which indicates the CSFs are valid. Figure 2 depicts the contrast sensitivity functions obtained from the younger child with brain injury (left), older child with brain injury (center), and healthy child (right). The low rate of false positives in the tracking-based task allows us to infer thresholds from the best results across multiple sessions with high confidence.
Conclusions :
Our findings indicate that a non-verbal gaze-based tracking task is suitable for measuring contrast sensitivity in both healthy and impaired populations. Due to the task’s inherent low false positive rate, the false negatives that often occur from inattention when testing impaired individuals can be discounted. Quantitative contrast sensitivity measurements in wider populations would be of great potential value to both basic and clinical vision science, and we suggest that gaze-based tracking may provide a means for assessing them.
This is a 2020 ARVO Annual Meeting abstract.