Eye, Brain, and Vision

your attentions misunderstood! The process of making visual saccades to items
of interest, in order to get their images on the fovea, is carried out largely by
the superior colliculus, as Peter Schiller at MIT showed in an impressive series
of papers in the 1970s.
The second set of facts about how we see is even more counterintuitive.
When we look at a stationary scene by fixating on some point of interest, our
eyes lock onto that point, as just described, but the locking is not absolute.
Despite any efforts we may make, the eyes do not hold perfectly still but make
constant tiny movements called microsaccades; these occur several times per
second and are more or less random in direction and about1 to 2 minutes of
arc in amplitude. In 1952 Lorrin Riggs and Floyd Ratliff, at Brown University,
and R. W. Ditchburn and B. L. Ginsborg, at Reading University, simultane-
ously and independently found that if an image is optically artificially stabi-
lized on the retina, eliminating any movement relative to the retina, vision
fades away after about a second and the scene becomes quite blank! (The
simplest way of stabilizing is to attach a tiny spotlight to a contact lens; as the
eye moves, the spot moves too, and quickly fades.) Artificially moving the
image on the retina, even by a tiny amount, causes the spot to reappear at
once. Evidently, microsaccades are necessary for us to continue to see station-
ary objects. It is as if the visual system, after going to the trouble to make
movement a powerful stimulus—wiring up cells so as to be insensitive to
stationary objects—had then to invent microsaccades to make stationary ob-
jects visible.
We can guess that cortical complex cells, with their very high sensitivity to
movement, are involved in this process. Directional selectivity is probably not
involved, because microsaccadic movements are apparently random in direc-
tion. On the other hand, directional selectivity would seem very useful for de-
tecting movements ot objects against a stationary background, by telling us
that a movement is taking place and in what direction. To follow a moving
object against a stationary background, we have to lock onto the object and
track it with our eyes; the rest of the scene then slips across the retina, an event
that otherwise occurs only rarely. Such slippage, with every contour in the
scene moving across the retina, must produce a tremendous storm of activity
in our cortex.

END STOPPING
One additional kind of specificity occurs prominently in the striate
cortex. An ordinary simple or complex cell usually shows length summation:
the longer the stimulus line, the better is the response, until the line is as long
as the receptive field; making the line still longer has no effect. For an end-

stopped cell, lengthening the line improves the response up to some limit, but
exceeding that limit in one or both directions results in a weaker response, as
shown in the bottom diagram on the facing page. Some cells, which we call
completely end stopped, do not respond at all to a long line. We call the region
from which responses are evoked the activating region and speak of the regions
at one or both ends as inhibitory. The total receptive field is consequently
made up of the activating region and the inhibitory region or regions at the
ends. The stimulus orientation that best evokes excitation in the activating
region evokes maximal inhibition in the outlying area(s). This can be shown
by repeatedly stimulating the activating region with an optimally oriented line
of optimal length while testing the outlying region with lines of varying orien-
tation, as shown in the top diagram on this page.
We originally thought that such cells represented a stage one step beyond
complex cells in the hierarchy. In the simplest scheme for elaborating such a
cell, the cell would be excited by one or a few ordinary complex cells with
fields in the activating region and would be inhibited by complex cells with
similarly oriented fields situated in the outlying regions. I have illustrated this
scheme in the bottom diagram on the preceding page. A second possibility is
that the cell receives excitatory input from cells with small fields, marked (a) in
the diagram on this page, and inhibition from cells with large fields, marked
(b); we assume that the cells supplying inhibition are maximally excited by
long slits but poorly excited by short ones. This second possibility (analogous
to the model for center-surround cells) is one of the
few circuits for which we have some evidence. Charles Gilbert, at Rockefeller
University in New York, has shown that complex cells in layer 6 of the mon-
key striate cortex have just the right properties for supplying this inhibition
and, furthermore, that disabling these cells by local injections causes end-
stopped cells in the upper layers to lose the end inhibition.

After these models were originally proposed, Geoffery Henry, in Canberra,
Australia, discovered end-stopped simple cells, presumably with the recep-
tive-field arrangement shown in the lower margin of this page. For such a cell,
the wiring would be analogous to our first diagram, except that the input
would be from simple rather than from complex cells. Complex end-stopped
cells could thus arise by excitatory input from one set of complex cells and