Synopsis

Rapid Eye Movements Enhance Information Acquisition

• Physics 19, s50
A model captures how the retina avoids tuning out during a fixed gaze.
Gustave Courbet/Public domain/Wikimedia Commons

Tiny, small-scale eye movements persist even when a human stares at a fixed point. Physiologists have long speculated about how these fixational eye movements, or “drift,” might help visual processing. Alexander Houston of the University of Glasgow in the UK and his collaborators now present a model that describes how both the stimulus—the visual scene—and the rapid eye movements affect visual performance [1]. They show how seemingly random eye movements serve to couple the spatial structure of a stimulus to a time-dependent visual response, with regimes that can be beneficial, detrimental, or ineffectual to information acquisition.

When you stare at an image, light travels through the lens in your eye before reaching the retina: the neural structure at the back of the eye that contains the photoreceptor array. Although the image appears clearly, if you stare fixedly for long enough, parts of it may fade from view. The retina “adapts” and stops signaling. Because of drift, however, each photoreceptor’s position shifts along a diffusive trajectory. In the model developed by Houston and his collaborators, these retinal movements impart a time dependence to spatial variations in the incoming light, overcoming the retina’s tendency to stop signaling.

After accounting for blur and other noise sources, the researchers derived an expression for “how much” signal the photoreceptor array captures in a given time. The value depends explicitly on the spatial scale of the stimulus, how long it takes the eye to adapt to a fixed image, and the diffusion constant of the eye’s movements. Additional analyses showed that information capture is maximized when the spatial scale matches how far the photoreceptors diffuse in the adaptation time. Excessively fast eye movements, on the other hand, have a detrimental effect. What’s more, the model suggests that measurements of visual performance previously attributed to complex neural processing can instead be understood simply as a consequence of the drifting motion of the eye.

–Rachel Berkowitz

Rachel Berkowitz is a Corresponding Editor for Physics Magazine based in Vancouver, Canada.

References

  1. A. J. H. Houston et al., “Information, movement, and adaptation in human vision,” Phys. Rev. Res. 8, 023083 (2026).

Subject Areas

Biological Physics

Related Articles

A Little Noise Helps a Soft Valve Close
Biological Physics

A Little Noise Helps a Soft Valve Close

A simplified model of the heart’s mitral valve shows that flow fluctuations can cause a soft valve to close at about one-tenth the pressure required under steady flow. Read More »

Cells Put a Price Tag on Sensing Their World
Biological Physics

Cells Put a Price Tag on Sensing Their World

A new model shows how cells could optimize biochemical sensing by balancing information gained against energy spent. Read More »

Nonreciprocity Sends Flocks into Chaos
Biological Physics

Nonreciprocity Sends Flocks into Chaos

Two intermingled species of active matter can exhibit coherent rotation or disorderly scrambling depending on their mutual interactions. Read More »

More Articles