Electrophysiology of Natural and Artificial Vision - Artificial Sight: Basic Research, Biomedical Engineering, and Clinical Advances - page 360

potential recordings is improved by the use of a large number of electrodes,
but is ultimately limited by the small contribution to the measured potentials by
individual neurons. Electrophysiological functional mapping has recently been
combined with structural imaging techniques that have high spatial resolution but
poor temporal resolution (e.g. MRI, CT); these multi-modal imaging strategies
exploit the strengths of each approach. The technology that allows us to monitor
thousands of individual neurons, in real time, in vivo, has yet to be developed.
In the meantime, we have several techniques that provide a great deal of infor-
mation about the nervous system. The following section provides an overview of
approaches used specifically to measure the response of the retina to stimulation
with light (i.e. photic stimulation).

Non-Invasive Techniques
The earliest technique used to record from the retina was the corneal
electroretinogram (ERG) [3]. In its simplest form, this is a straightforward surface
potential measurement, where two electrodes are placed on the body surface near
the eye, and a potential difference that varies with time is recorded following the
delivery of a light stimulus. Typically, one electrode is placed in contact with
the cornea of the eye itself, and a second reference electrode is placed on the
skin a few inches away (on the temple, forehead, or earlobe).
Almost all techniques for electrophysiological recording, including ERG
recording, employ a differential amplifier, in which the potential difference
between a reference electrode and ground electrode is effectively subtracted
from the potential difference between the recording electrode and the same
ground electrode. Ideally, the potential on the reference electrode would reflect
only contributions from noise sources that are also picked up by the recording
electrode. Thus, the signal downstream of the differential amplifier consists of
only the desired signal:

Signal + Noise − Noise = Signal

Recording Reference Amplifier

Electrode Electrode Output

The degree to which this strategy works can be quantified by the common
mode rejection ratio (CMRR), which is the ratio of the differential mode gain
(DMG, the gain applied to the potentials not appearing on both electrodes) to
the common mode gain (CMG, the gain applied to the potentials common to
both electrodes).

CMRR = DMG/CMG

The CMG is ideally zero, but is typically close to one in real amplifiers; this
is much smaller than the DMG of 100-1000 typically applied to field potential
signals. A CMRR of 100 or better is considered quite good.