Electrophysiology of Natural and Artificial Vision - Artificial Sight: Basic Research, Biomedical Engineering, and Clinical Advances - page 363

the a-wave induced by the bright flash, A sat t, is a measure of how far the
photoreceptors were from saturation at the time the bright flash was delivered.
Performing the subtraction A sat − A sat t reveals how far the photoreceptors were

from saturation due to the test flash at the time the bright flash was delivered,
At =

A sat t. By delivering the bright flash at several different times after
the test flash, in several paired-flash trials, A(t) can be calculated for the full
time course of the photoreceptor response. An important feature of this approach
is that it is used to isolate the response of the first-order neurons contributing to
the corneal ERG, and this is the property of the pfERG technique that may be
exploited in studying electrical stimulation.
The following two techniques are common in clinical electrophysiology of
the eye. Because they exploit network properties of the retina, their application
to studying the response of a diseased retina to electrical stimulation is much
less straightforward than the fERG and the pfERG. They are described briefly
here as an illustration of the variety of ERG techniques available, and with the
hope that once we know more about the physiology of the degenerate retina,
correlates may be developed for use in studying electrical stimulation.
The pattern ERG is primarily used to measure the activity of ganglion cells.
The third-order neurons of the retina are most sensitive to spatial differences
in retinal illumination. The stimulus of the pERG is therefore a high-contrast
checkerboard pattern subtending the central visual field that inverts (black
becomes white, white becomes black) at a regular frequency (1-10 Hz). The mean
luminance of the checkerboard remains constant, which minimizes the influence
of the photoreceptors and second-order neurons to the recorded response. Like
the pfERG, the pERG can provide information about a specific class of cells in
the retina, but does not convey information about the location of a focal retinal
anomaly. That is, the recorded response is still the summed response of contri-
butions from all areas of the retina subtended by the stimulus pattern (typically
the macular region).
The multifocal ERG is a technique used to perform functional mapping of the
retina, and can provide information related to cell type and spatial location [11].
However, in contrast to most functional mapping methods that use multiple
recording electrodes, the mfERG uses a single corneal electrode. The goal is to
produce a map of the retina that indicates the strength of the local response to a
given stimulus. This is accomplished by stimulating different areas of the retina,
and then correlating the area in visual space associated with the stimulus to an
area on the retina. Imagine viewing a large screen divided into two halves, upper
and lower. If the upper half is bright and the lower half is dark, the inverted image
of the screen will illuminate the inferior half of the retina, while the superior
half of the retina remains in the dark. The signal recorded at the cornea would
be the response originating in the stimulated inferior region of the retina; the
superior region would not contribute. If the screen was then divided into, say, a
10

A sat −

10 array of pixels, the response to each box could be mapped to one of one
hundred regions in the retina. In practice, this method must account for the mean
illuminance of the retina, which determines the state of light adaptation, and

×