Scribd
Upload
55 views1 page

Field ), I /I'RB LB Bockground

The document discusses visual adaptation and retinal gain controls. It explains how retinal neurons can detect contrast through temporal modulation caused by eye movements, and how receptive field surrounds make neurons more sensitive to patterns with narrow spatial gradients. It also discusses how brightness constancy relies on local contrast near object borders rather than comparing light across the entire retina.

Uploaded by

Akica
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
55 views1 page

Field ), I /I'RB LB Bockground

The document discusses visual adaptation and retinal gain controls. It explains how retinal neurons can detect contrast through temporal modulation caused by eye movements, and how receptive field surrounds make neurons more sensitive to patterns with narrow spatial gradients. It also discusses how brightness constancy relies on local contrast near object borders rather than comparing light across the entire retina.

Uploaded by

Akica
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

VISUAL ADAPTATION AND RETINAL GAIN CONTROLS 271

(a) expect deviations from brightness constancy.


/I-Ro = Lo
Receptive~ This line of reasoning leads to the conclusion that
Field~),i~ /I'RB=LB the detection of contrast depends on light
adaptation. A different view is that the dependence
Bockground of retinal responses on contrast is a consequence
of the spatially antagonistic c e n t e r - s u r r o u n d
organization of receptive fields (discussed below in
Appendix 2), but we believe this view to be
incorrect. If there are eye movements, spatial
contrast will be " s e e n " by the highly localized
(b) Receptive receptive fields of retinal neurons as successive
Field,) contrast, or in other words as temporal modulation
of the amount of light falling on those receptive
Background fields (Helmholtz, 1909). As shown above, in this
situation, the mechanisms of adaptation together
with the compact nature of the receptive fields
FIG. 3. The invariance of contrast with changes in level of produce a response dependent on contrast without
illumination, and how this is sensed by a visual cell's receptive any requirement for a spatially antagonistic
field. This picture shows the luminance profile of an object surround. The purpose of having receptive field
(with reflectance R o) on a background (with reflectance RB)
when both are illuminated uniformly, with illumination L
surrounds probably is to make retinal neurons more
The position of the receptive field of an individual retinal sensitive to patterns with narrow spatial gradients,
cell is indicated by the shaded rectangle. In (a), the visual or high spatial frequencies, than to coarse patterns
cell is "looking" at the background. In Co), an eye movement
has carried the receptive field of the cell onto the object.
or diffuse light. Put another way, with light
Equation (5) in the text demonstrates that this cell will respond adaptation but without receptive field surrounds,
to the Weber contrast of the object on the background, i.e. retinal neurons would respond to contrast.
(Ro-RB)/RB, which is independent of illumination L As However, with receptive field surrounds but without
explained in the text, in order for this calculation to work
the cell must be influenced by a gain control which adjusts adaptation, the retina would be unable to respond
the cell's gain to be reciprocal with background illumination. to contrast; it would saturate quickly (see Section
1.2.2.). This point of view is similar to that put
forward by Whittle and Challands (1969).
proportional to the reciprocal of the value of the
illumination at which the cell's receptive field has Consideration of Mach's statement together with
been " l o o k i n g " up to the moment it crosses the more recent work on brightness constancy under
border. This inference is suggested by Weber's Law, conditions of non-uniform illumination leads to an
equation (2). Thus the signal in our hypothetical interesting conclusion about the spatial spread of
n e u r o n will equal the stimulus I'Ro-I'RB adaptation. Mach, and others after him (Hering,
multiplied by the gain, K/(I. RB), where K is a pro- 1920; Helson, 1964), have supposed that the eye
portionality constant. The conclusion is that what compared, "the quantity of light on the entire retina
the visual system uses to characterize an object is and the image of the paper", in order to calculate
a signal the brightness of the paper. This is not correct. The
first correction is that brightness depends on a local
S = K" (I'Ro-I" RB)/I" RB (5) phenomenon, the contrast near the border of an
object, not on the quantity of light in the image of
which is proportional to the Weber contrast. In this the object (see Fig. 2). Second, the retina does not
case the neural response to the object is invariant compare the quantity of light " o n the entire retina"
with respect to changes in the illumination, I. This with the amount of light in the object, as Mach
chain of reasoning only holds if the gain of the suggested. As Davidson and Freeman (1965) and
retina is reciprocal with background illumination, Land and McCann (1971) have shown, brightness
in analogy with Weber's Law. When retinal gain is constant under conditions of spatially non-
is not reciprocal with background, one should uniform illumination. We have argued that retinal

You might also like