There are
ganglion cells in the
retina of certain
animals (though not
humans) which are responsive to
light only when the light
stimulus is moving in a certain direction. The mechanism of such cells is a
mind-boggling feat of
evolution.
I have tried to illustrate with a diagram as best I can:
LIGHT
|\/\/\/\/|
|/\/\/\/\|
|\/\/\/\/|
__|/\/\/\/\|__
\\/\/\/\/\/\//
\\/\/\/\/\//
\\/\/\/\//
\\/\/\//
\\/\//
\\//
\/
---------Null Direction--------->
<------Preferred Direction-------
Photoreceptor Cells
/ | \
(___A___) (___B___) (___C___)
+ - + - +
+ - + - +
+ - + - +
+ - + - +
+ - + - +
+ - + - +
+ - + +
+ + +
+ + +
(Ganglion Cell)
||
||
||
||
VV
To optic nerve
The cell that mediates the coding of the direction of stimulus movement is the ganglion cell. It receives input from a certain number of photoreceptor cells on the retina. In the above diagram and the following description, I have limited the number of photoreceptor cells providing input to the ganglion cell to three for simplicity, but really, there are many more.
So how does the ganglion cell respond only to movement in a particular direction?
The mechanism involves a delayed inhibition system. Using the diagram above, if a light stimulus excites receptor cell A, then the cell sends an excitatory signal to the ganglion cell (excitatory signals are represented by the “+” lines). It also sends a delayed inhibitory signal to cell B (inhibitory signals are represented by the “–“ lines). When the moving light stimulus has moved across to stimulate cell B, it causes cell B to also send an excitatory signal to the ganglion cell. However, we find that the time that the light reaches B coincides with the time the delayed inhibition from A reaches B, thus the excitatory signal is cancelled out by A’s inhibition. At B, a further inhibitory signal is sent to cell C to preclude cell firing when light reaches there.
So when the light stimulus moves from A to B to C, there is no net excitatory signal reaching the ganglion cell, and so, the ganglion cell does not respond. The direction is called the “null” direction. However, if the stimulus was to move from C to B to A, then the moving light stimulus always precedes the inhibition it leaves behind. This is called the “preferred” direction. Therefore, in the preferred direction, the ganglion cell receives a constant excitatory input and will fire continuously. This is how a ganglion cell can selectively respond to a light stimulus depending upon its direction of movement.
But a directionally selective ganglion cell cannot signal the exact direction of a moving light stimulus on its own, since it will respond to some degree when moving in directions between the null and preferred direction. This is because the inhibitory signals do not exactly follow the path of light direction and some excitation finds its way to the ganglion cell. The problem is solved by looking at the responses of a population of these cells, each attuned to be maximally sensitive to a different direction. Such a population will indicate the exact direction through analysis of the net response of the group.