I had read some information at a sight about sensation and perception some
time ago. I found the site again,
http://hyperphysics.phy-astr.gsu.edu/hbase/vision/rodcone.html It has a very
nice graph of the response of rhodopsin, it simply does not respond to light
of greater than 600nm. This allows you to use a red flash light if the
light does not go over 600nm with no effect on your rods. I can see why
LED's would be popular due to the "purity" of the wavelength. I guess dim
lights would last longer.

The cones seem to turn off more when there is insuffient light in order to
stop from fuzzing the picture, like a bad cable connection with lots of
sparkles or more closely to our subject area, like a CCD not properly
cooled. What's a real bit and what's a fake one? There are circuits in the
retina that compare light levels for cones and suppress the activity if the
"vote" is low light. BUT there is no circuit between rods and cones
specifically! Red cones can be used without any real inhibitory effect on
rods. There is some "ganging" done in rods and in cones, but again nothing
between the two cell types, to my knowledge.

Interestingly, our red cones are OPTIMALLY responsive to reddish-orange
light. Our rods are almost nonresponsive to the same. About 64% of our
cones are "red" which is probably good also. I have read an article on the
AAAA site saying that green light was better because you could use lower
light, but this does not seem to make sense. First the usable areas under
the curve for red vs green seem to be equal, but most of red's does not
overlap the rod's curve of sensitivity. Further there are twice as many red
receptors. Nothing was presented that showed that the Green receptors are
twice as sensitive. The curve of sensitvity of Green and Red do over lap
quite a bit, but the luminence must be less due to the over lap with the
rod's. The rod's are over twice as efficient in their use of light at
500nm, so any move toward the green would seem counterproductive.

Most of the research comes from the Navies of the world in the 1920's and
1930's through the present. Red lighting was discovered to allow search
positions to see that iceberg while any other light shut good vision down.
This was basic "Edision" style experimentation, no understanding of the
mechanism. Later specialists tracked down rhodopsin's unwinding
wavelengths. The use of orange-red lighting in car dashes is due to recent
research.

-----Original Message-----
peter erdman
Sent: Sunday, August 26, 2001 8:08 PM
Subject: Re: [M]: A Night Vision Question


A couple of weeks ago, a question came up on the proper color for a
supplemental light that would still preserve one's night vision.
Conventional wisdom dictates dim red light, but I've pursued this a bit to
find an explanation. I haven't found this conclusion written anywhere in
my sources, but then I didn't get too carried away with researching the
topic either. Usually when I "discover" something it later turns out that
it has been well known for 50 years or so.

One of the mechanisms for dark adaption is the increase in the pigment
rhodopsin in the eye's rods. It is the rods which are finally the
effective sensors at low light levels. The dark adapted eye's peak
sensitivity shifts to shorter wavelengths with a peak near 0.5 microns
instead of the peak (for "high" light levels) at ~0.55 microns. The
sensitivity vs wavelength curve of the dark adapted eye (scotopic vision)
matches the absorption cross section curve for rhodopsin--which is very low
at the longer wavelengths.

Other things are also going on as the eye adapts to the dark, and that
makes things more complex. Neural pathways are modified to effectively
interconnect many rods to a single nerve ganglia. This is one of the
reasons why a larger object (on the retina) is more easily detected at low
light levels than a smaller one of equal surface brightness. There are
more rods off the optics axis of the eye than on--that's why "averted
vision" works. Etc.

It was the absorption curve of rhodopsin that intrigued me, and begged the
same question of whether the important factor was the dimness of the light
or its wavelength? If the dark adapted eye is less sensitive in the red,
wouldn't it just mean that a brighter light is required than if the light
had a wavelength at the peak sensitivity? The detection of light in the
dark adapted eye involves the absorption by rhodopsin and destroys the
particular molecule in the process (dissociation). Too much light and the
rhodopsin level decreases and night vision is destroyed. Hence, if the
probability of absorption at longer wavelengths is less, then a brighter
light is needed. A dimmer light could be used at the peak. Both would
result in the destruction of the same number of rhodopsin molecules, and
therefore have the same effect on the night vision?

This was the question that I pursued--did wavelength matter? Standard
answer I always got was the usual "that's what everybody uses so it must be
right". Not too convincing an argument.

So here's my explanation of the process. Dark adaption increases the eye's
sensitivity at the shorter wavelengths, but not much in the red (rhodopsin
doesn't absorb well there). Since rhodopsin doesn't absorb well in the
red, red light will have little ability to destroy the increased
concentration of rhodopsin in the dark adapted eye. Emphasis here must be
placed on lower absorption, not zero. Bright red will still cause
problems. This is still the conventional wisdom.

So how do we see red light with the dark adapted eye? The other sensors in
the eye, the cones, are usually ineffective at low light levels since their
sensitivity is little enhanced, but they continue to function--and they
have at least the same sensitivity to red light that they had before the
eye became adapted. So when we turn on the red light, we are actually
using our cones to detect it and the dark adapted rods don't detect the red
light (as long as it is dim enough). The entire process depends on the
slight difference between being bright enough to be able to see the red
light with the cones, but not bright enough to destroy rhodopsin in the
dark adapted rods.

So I did a quicky, unscientific, unquantitative experiment. How bright
does my red LED flashlight appear to my adapted and unadapted eye? If it
is indeed only the cones that are detecting the light, then there will be
little or no change in the flashlight's apparent brightness. I did the
test, and since I knew what result I should get, of course that's exactly
what I perceived. Whether it's right or not is another matter. Your
mileage may vary.

Oh yes, I did the observation through a 12" LX-200.

Peter

View index by [date] [author] [subject]
Previous message: Re: [M]: Best Pair, Paul Rodman
Next message: [M]: Re: Focal Reducer Strangeness..., Bostjan
Previous message in thread: Re: [M]: A Night Vision Question, Bostjan