Saccades and Microsaccades.

I. Introduction.

A saccade is a fast movement of an eye, head or other part of an animal's body or device. It can also be a fast shift in frequency of an emitted signal or other quick change. However, this article deals with saccadic eye motion.

Eye saccades are quick, simultaneous movements of both eyes in the same direction. Initiated by eye fields in the frontal and parietal lobes of the brain, saccades serve as a mechanism for fixation, rapid eye movement and the fast phase of optokinetic nystagmus. The word appears to have been coined in the 1880s by French ophthalmologist Émile Javal, who used a mirror on one side of a page to observe eye movement in silent reading, and found that it involves a succession of discontinuous individual movements.

Humans and other animals do not look at a scene in a steady way. Instead, the eyes move around, locating interesting parts of the scene and building up a mental 'map' corresponding to the scene. One reason for saccades of the human eye is that the central part of the retina, the fovea, plays a critical role in resolving objects. By moving the eye so that small parts of a scene can be sensed with greater resolution, body resources can be used more efficiently.

In addition, the human eye is in a constant state of vibration, oscillating back and forth at a rate of about 30-70 Hz. These microsaccades are tiny movements, roughly 20 arcseconds in excursion and are completely imperceptible under normal circumstances. They serve to refresh the image being cast onto the rod cells and cone cells at the back of the eye. Without microsaccades, staring fixedly at something would cause the vision to cease after a few seconds since rods and cones only respond to a change in luminance.

When the brain is led to believe that the saccades it is generating are too large or too small (by an experimental manipulation in which a saccade-target steps back or forward, contingent on the eye movement made to acquire it), saccade amplitude gradually decreases (or increases), an adaptation (also termed gain adaptation) widely seen as a simple form of motor learning, possibly driven by an effort to correct visual error.

This effect was first discovered in humans with ocular muscle weakness brought on by disease or tenectomy. In these cases, it was noticed that the patients would make hypometric (small) saccades with the affected eye, and that they were able to correct these errors over time. This led to the realization that visual error (the difference between the intended past-saccadic point of regard and the target position) played a role in the homeostatic regulation of accurate saccades. Since then, much scientific research has been devoted to various experiments employing saccade adaptation.

II. Kinematics of Saccades.

Saccades are the fastest movements produced by the human body. The peak angular speed of the eye during a saccade reaches up to 1000°/sec in monkeys (somewhat less in humans). Saccades to an unexpected stimulus normally take about 200 milliseconds to initiate and then last from about 20 to 200 milliseconds, depending on their amplitude. Under certain laboratory circumstances the latency of saccade production can be cut nearly in half (express saccades).

The amplitude of a saccade is the angular distance that the eye travels during the movement. For amplitudes up to about 60 degrees, the velocity of a saccade linearly depends on the amplitude (the so called "saccadic main sequence"). For instance, a 10° amplitude is associated with a velocity of 300°/sec, and 30° is associated with 500°/sec. In saccades larger than 60 degrees, the peak velocity starts to plateau (non-linearly) toward the maximum velocity attainable by the eye.

Saccades may rotate the eyes horizontally or vertically, or in any oblique direction to change gaze direction (the direction of sight that corresponds to the fovea), but normally saccades do not rotate the eyes torsionally. Torsion can be defined as clockwise or counterclockwise rotation around the line of sight when the eye is at its central primary position. Defined this way, Listing's law says that when the head is motionless, torsion is kept at zero.

Head-fixed saccades can have amplitudes of up to 90° (from one edge of the oculomotor range to the other), but in normal conditions saccades are far smaller, and any shift of gaze larger than about 20° is accompanied by a head movement. During such gaze saccades, first the eye produces a saccade to get gaze on target, whereas the head follows more slowly and the vestibulo-ocular reflex causes the eyes to roll back in the head to keep gaze on the target. During these head movements Listing's law is no longer obeyed.

Saccades, as well as microsaccades, can be distinguished from other eye movements (ocular tremor, ocular drift, smooth pursuit) using their ballistic nature: their top velocity is proportional to their length. This property can be used in algorithms for saccade detection in eye tracking data.

III. Types of Saccades.

Saccades are measured or investigated in four ways:

• In a visually guided saccade, an observer's eyes move towards a visual onset, or stimulus. This is typically included as a baseline when measuring other types of saccades.

• In an antisaccade, an observer's eyes move away from the visual onset. They are more delayed than visually guided saccades, and observers often make erroneous saccades in the wrong direction. A successful antisaccade requires inhibiting a reflexive saccade to the onset location, and voluntarily moving the eye in the other direction.

• In a memory guided saccade, an observer's eyes move towards a remembered point, with no visual stimulus.

• In a sequence of predictive saccades, an observer's eyes are kept on an object moving in a temporally and/or spatially predictive manner. In this instance, saccades often coincide (or anticipate) the regularly moving object.
  

IV-A. Saccadic masking.

It is a common but false belief that during the saccade, no information is passed through the optic nerve to the brain. Whereas low spatial frequencies (the 'fuzzier' parts) are attenuated, higher spatial frequencies (an image's fine details) which would otherwise be blurred out by the eye movement remain unaffected. This phenomenon, known as saccadic masking or saccadic suppression, is known to occur in the time preceding a saccadic eye movement, implying neurological reasons for the effect, rather than simply the image's motion blur.

A person may observe the saccadic masking effect by standing in front of a mirror and looking from one eye to the next (and vice versa). The subject will not experience any movement of the eyes nor any evidence that the optic nerve has momentarily ceased transmitting. Due to saccadic masking, the eye/brain system not only hides the eye movements from the individual but also hides the evidence that anything has been hidden. Of course, a second observer watching the experiment will see the subject's eyes moving back and forth. The function's main purpose is to prevent smearing of the image.

IV-B. Spatial Updating.

When a visual stimulus is seen before a saccade, subjects are still able to make another saccade back to that image, even if it is no longer visible. This shows that the brain is somehow able to take into account the intervening eye movement. It is thought that the brain does this by temporarily recording a copy of the command for the eye movement, and comparing this to the remembered image of the target. This is called spatial updating. Neurophysiologists who have recorded from cortical areas for saccades during spatial updating have found that memory related signals get remapped during each saccade.

IV-C. Trans-saccadic Perception.

It is also thought that perceptual memory is updated during saccades so that information gathered across fixations can be compared and synthesized. However, the entire visual image is not updated during each saccade, only 3-4 features or objects if they are attended to. Some scientists believe that this is the same as visual working memory, but as in spatial updating the eye movement has to be accounted for. The process of retaining information across a saccade is called trans-saccadic memory and the process of integrating information from more than one fixation is called trans-saccadic integration.

V. Recent Developments.

A recent study by researchers at the Barrow Neurological Institute led by Susana Martinez-Conde has taken an important step toward understanding how the brain uses saccades and microsaccades in order to "sharpen" a scene. Previously, it has been unclear whether saccades and microsaccades have inherent differences or not. Here, the researchers found that both movements are likely generated by the same neural mechanism in the brain's strategy for optimal visual sampling.

In experiments, participants viewed various visual scenes ranging from blank images to complex pages of the Where's Waldo? books by Martin Handford. Then, the researchers measured the amount of saccades and microsaccades produced by the eyes when participants were either fixating on a specific point in an image or freely viewing the entire image.

Because microsaccades are operationally defined as movements that occur when fixating on a scene, the researchers looked for the same small magnitude of these movements when participants were freely viewing the scene. The researchers defined saccadic movements relative to microsaccades, with saccades having higher magnitudes than microsaccades.

The results showed that, when participants were freely viewing images, they produced more microsaccades when looking at the complex scenes than when viewing the blank and duller scenes. Specifically, more microsaccades occurred when participants were viewing an object of interest, such as when they found Waldo. Since participants stared longer at fixed points in the blank scenes than in the Waldo scenes, the increase in microsaccades could not be attributed to viewers fixating on Waldo for long times.

Instead, as the scientists explained, these results may support the proposal that microsaccades significantly re-sharpen an image and improve spatial resolution, as suggested in a recent study. While microsaccades occurred when participants viewed target objects, saccades occurred more often when participants freely viewed complex images as a whole. But rather than differentiating between microsaccades and saccades, the researchers suggested that both movements belong on a continuum of eye movements, which may together reflect an optimal sampling method by which the brain discretely acquires visual information.

The researchers hope that these findings may help understand the neural mechanisms underlying search behavior, both in the normal brain and in patients with eye movement deficits. In addition, understanding saccades and microsaccades could also help researchers design future neural prosthetics for patients with brain damage, as well as help to create intelligent machines that can see as well as humans.

Source: The Wiki article, PhysOrg and Journal of Vision. (Abstract, PDF)