Estimation of Contrast Sensitivity From Fixational Eye Movements

Area of Science:

• Ophthalmology and Vision Science
• Neuroscience
• Biomedical Engineering

Background:

• Microsaccades are small, involuntary eye movements occurring even during fixation.
• Microsaccade rate is influenced by the presentation of salient visual stimuli.
• Contrast sensitivity, a key measure of visual function, varies across populations.

Purpose of the Study:

• To develop a practical, objective, and robust method for estimating contrast sensitivity.
• To utilize microsaccade rates as a proxy for contrast sensitivity estimation.
• To validate the method across a diverse population including visually normal, amblyopic, and cataract patients.

Main Methods:

• Recorded binocular eye movements (microsaccades) during monocular contrast sensitivity testing.
• Employed a probabilistic inference approach to link microsaccade rates to stimulus presence.
• Estimated contrast sensitivity based on the scaled log-likelihood ratio of microsaccade rates across varying contrasts.

Main Results:

• Contrast sensitivity was successfully estimated from microsaccade rates in all participants.
• Microsaccade-derived contrast sensitivity estimates were strongly correlated with behavioral measures (Spearman's ρ = 0.74).
• Estimates showed minimal bias (1.2% mean difference) but lower precision than behavioral methods, requiring ~20x more presentations.

Conclusions:

• Microsaccade rate signatures provide a robust basis for contrast sensitivity estimation via probabilistic inference.
• Heterogeneity in microsaccade signatures across diverse populations does not preclude estimation.
• Further refinement is needed to improve the precision of microsaccade-based contrast sensitivity measurement.