Reading 6: Control of Saccadic Eye Movements

Question 1

Why do we understand the systems that control eye movements better than any other movement control systems?

Answer 1

1) Eye movements are comparatively simple; the eye is a sphere, controlled by 6 muscles that move it in its 3 axes of rotation. Compared to your arms, for example, arm movements involve coordination of 3 limb segments and 3 joints, each capable of an enormously complex range of movements.
2) The load on the eyes is always the same. This is different from your arms, in which the loads vary greatly as you pick up , hold and move objects of different weights.
3) brainstem neurons responsible for eye movements are known and have been extensively studied using electrophysiological recording.

Question 2

What are the different systems controlling eye movements?

Answer 2

Eye movements are controlled by five separate systems. 1) The optokinetic and vestibulo-ocular systems enable objects of interest to stay focused on the fovea even though your head is constantly moving. (These systems enable you to read a book on a bumpy bus ride.)
2) The vergence, smooth pursuit and saccade systems enable you to direct vision toward objects of interest and follow the objects as they move.

Question 3

Saccade system

Answer 3

This system is responsible for the rapid eye movements that we use to visually scan objects and to shift visual attention from one thing to another.

Question 4

Circuitry of the saccades

Answer 4

• The circuitry for saccades involves regions of the cerebral cortex, including the lateral intraparietal area (Area LIP), the frontal eye fields, and the prefrontal cortex, as well as the basal ganglia and nuclei in the brainstem
• The superior colliculi are key hubs connecting eye movement commands from the cortex to the eye movement circuitry in the brainstem.

Question 5

Brainstem

Answer 5

• The brainstem is where the basic motor circuitry resides.
• Consider an experiment in which a monkey is staring straight ahead and then makes a horizontal saccade to the right. Saccades are very fast so there is no time for sensory feedback during the saccade. Therefore, the saccade circuitry must encode, in a feedforward manner, the amplitude of the saccade (i.e., how fast and far the eye moves) and the new position of the eye.
• Furthermore, saccades involve conjugate movements of both eyes, so we will also need to consider how the circuitry enables the two eyes to move together.

Question 6

Muscles responsible for saccades

Answer 6

• A saccade to the right requires simultaneous contraction of the right lateral rectus and the left medial rectus.
• This coordinated horizontal movement of the eyes occurs because of two groups of neurons in the abducens nucleus: 1) motor neurons that activate the ipsilateral lateral rectus through the abducens nerve, and 2) interneurons that cross the midline to innervate motor neurons in the contralateral oculomotor nucleus. These oculomotor neurons activate the contralateral medial rectus.
• As a result, activation of neurons in the right abducens nucleus causes the ipsilateral lateral rectus and the contralateral medial rectus contract together, resulting in a rightward saccade.

Question 7

When the eye makes a saccade to the right, what is happening in the brainstem?

Answer 7

• When the eyes make a saccade to the right, motor neurons and interneurons in the right abducens nucleus fire a burst of action potentials, called a saccade pulse, that drives contraction of the ipsilateral lateral rectus and contralateral medial rectus during the saccade (Fig. 5).
• The frequency of the burst determines the velocity of the saccade.
• The burst is followed by a sustained increase in steady-state firing, called the saccade step, which encodes the new eye position (Fig. 5).
• These two different components of the response in the abducens nucleus are driven by separate groups of neurons in the brainstem.
• Burst neurons, located adjacent to the abducens nucleus, fire a burst of action potentials that drive the initial pulse of abducens neurons during the saccade (Fig. 5). Burst neurons are driven by excitatory inputs from the superior colliculus (see below). Their activity is also under inhibitory control from inhibitory omnipause neurons located near the midline of the brainstem. Omnipause neurons fire at steady rates between saccades, but stop firing during a saccade, thus relieving inhibition on the burst neurons, and enabling them to fire (Fig. 5).

Question 8

The new eye position is encoded by the step change in firing rate after the burst. But how is this new firing rate determined?

Answer 8

• David Robinson, whose research in the early 1960s made seminal contributions to our understanding of the saccade circuitry, predicted that the step change in firing rate could be set by a neural integrator that integrates (i.e., “adds up”) the saccade velocity signal from the burst neurons and converts it into a position signal.
• This position signal drives the new steady state firing of the abducens neurons. Robinson originally proposed the neural integrator on theoretical grounds, but it has since been shown to correspond to a specific network of neurons in the brainstem.
• These neurons integrate the burst and convert it into a new position signal (Fig. 5).

Question 9

Superior collicullus

Answer 9

• The superior colliculi (singular: colliculus) are layered structures that form two distinctive bumps on the dorsal surface of the midbrain (Fig. 6).
• The superficial layers are visual. They receive input from the retina and from visual areas of cortex and project back to visual cortex through the pulvinar nucleus of the thalamus. The superficial layers of each colliculus form a map of contralateral visual space.
• The intermediate and deep layers are involved in eye movements. They form a map of contralateral saccades of different amplitudes and directions.
• Horizontal movements of different amplitudes are mapped on the dorsal-ventral axis, whereas vertical movements are mapped on the medial-lateral axis (Fig. 7).
• The most rostral region of each colliculus represents the fovea. This region is called the fixation zone. It inhibits saccades when the eyes are fixed on an object of interest.

Question 10

Fixation neurons

Answer 10

• Between saccades, when the eyes are still, fixation neurons in the fixation zone are active (Fig. 7).
• When a saccade is initiated, neurons in the fixation zone stop firing, whereas saccade neurons in the region of the colliculus contralateral to the direction of movement and corresponding to the amplitude and direction of the saccade, fire a burst of action potentials (Fig. 7).
• At the end of the saccade these neurons stop firing and neurons in the fixation zone start to fire again. Thus, the neurons in the fixation zone look sort of like the omnipause neurons, and these neurons do indeed project to the omnipause neurons and contribute to control of their behavior.
• The colliculus neurons that fire during the saccade project to the burst neurons in the brainstem. Their firing patterns are reminiscent of the burst neurons, but there is a crucial difference. The amplitude and direction of saccades is encoded on the spatial map of the colliculi (Fig. 7).
• In contrast, saccade amplitude is encoded by the burst neurons as the frequency of their bursts.
• Thus, the transfer of the instructions for a saccade from the colliculus to the brainstem requires a spatial-temporal transformation.

Question 11

Cerebral cortex role in saccades

Answer 11

• Two important areas in cortical control of saccades are area LIP in the parietal lobe and the frontal eye field in the frontal lobe (Fig. 2).
• Area LIP receives visual input through the dorsal visual stream. It is involved in both saccades and visual attention.
• Lesions in this region impair saccades in response to bottom-up visual stimuli, but not to cognitively evoked saccades.
• Parietal lesions, especially in the right hemisphere also result in sensory neglect; patients ignore the side of the world contralateral to the lesion and don’t make saccades in that direction.

Question 12

Frontal eye fields and saccades

Answer 12

• The frontal eye fields are more directly connected to generation of saccades.
• Neurons in the frontal eye fields control colliculus neurons both directly, through projections to the colliculus, and indirectly through the basal ganglia.
• Lesions to the frontal eye fields disrupt cognitively driven saccades, but subjects are still able to make saccades in response to visual stimuli.

Question 13

Basal Ganglia and saccades

Answer 13

• The projection from the frontal eye fields to the basal ganglia provide a good example of how the basal ganglia control movement.
• In this case, the output nucleus is the substantia nigra pars reticulata rather than the globus pallidus internal segment, and the output projects to the superior colliculus rather than to the thalamus, but other than that, the mechanism is like what we will talk about in the basal ganglia lectures.
• When the eyes are not moving, inhibitory neurons in the substantia nigra fire spontaneously at high frequency, tonically inhibiting colliculus neurons.
• A saccade command from the frontal eye fields activates inhibitory neurons in the caudate nucleus that project to the substantia nigra.
• This activity inhibits the substantia nigra, thus relieving inhibition on the superior colliculus, enabling it to direct saccades (Fig. 8).