Abstract
We previously demonstrated that contrast sensitivity can be accurately predicted from changes in the rate at which normal individuals make microsaccades: small, involuntary fixational eye movements (Scholes et al. 2015, Proc. R. Soc. B 282: 20151568). Here, we expand this previous work to include two groups of individuals with visual sensitivity loss that has either an optical (cataract) or neural (amblyopia) origin. Contrast sensitivity functions were first measured psychophysically and used to inform the choice of stimulus parameters for the main experiment. The highest spatial frequency for which each subject could reliably detect a 75% contrast Gabor was selected and contrast levels were adjusted to span threshold. For the majority of trials, subjects were required to fixate a central dot as Gabor stimuli were passively presented. However, on a subset of trials (indicated by a tone) subjects were required to discriminate the orientation of the Gabor (±45 degrees), allowing us to collect a concurrent behavioural estimate of contrast sensitivity. Eye movements were recorded using an Eyelink 1000 and microsaccades were detected using a velocity-based algorithm. Each subject completed 400 passive trials per contrast and microsaccade rates were established in an epoch around stimulus presentation. Microsaccade rate modulations were characterized by computing the absolute difference between an individual's base microsaccade rate and the modulated rate for a particular contrast. Oculomotor thresholds, calculated from logistic fits to this measure, corresponded closely to behavioural thresholds, capturing the variability in absolute thresholds (11 – 64%) at a range of spatial frequencies (1-12 cpd) across 30 subjects. Our data indicate that visual sensitivity can be accurately predicted using fixational eye movements both for a range of stimulus conditions and in individuals with sensitivity losses with both optical and neural origins.
Meeting abstract presented at VSS 2017