Physical reality and perception

[From Bruce Abbott (980309.1400 EST)]

Bill Powers (980309.0906 MST) --

Bruce Abbott (980309.0900 EST)

In most (all?) cases perception is a construct from the start. The retina
contains four kinds of photoreceptor, three of which together provide the
basis of color vision. These three are "tuned" to different frequency bands
of the electromagnetic spectrum; their peaks overlap somewhat. What we call
yellow usually begins as a particular frequency of light, one that activates
the two lowest-frequency photoreceptors about equally well.

I agree in general with what you're saying, but disagree with a few
details. The experience of yellow, for example, would very seldom begin
with a single-frequency light stimulus. If you use a specroscope to look
at any natural object that we see as yellow, there is essentially no chance
that we would find all the light energy concentrated at a single wavelength.

For the purpose of _testing_ the visual system there are certainly ways to
generate monochromatic light. My point was that a particular frequency of
light (e.g., from sodium emission) may give rise under certain standardized
conditions to a particular perception of color (e.g., yellow) -- that one
can find a one-to-one correspondence. But even so, this is a perceptual
construction, as there are no "yellow" receptors in the system. Although we
can perceive yellow, we have no single receptors that are specifically
sensitive to that band of the spectrum.

The spectral bands of the eye overlap not slightly, but mostly, at least
for the red-green bands. Edwin Land showed that the most important
distinction in color vision is between long and short wavelengths; the same
object can appear to be different colors depending on the colors of other
objects in the same visual field -- yet a given object can be seen as the
same color even though it is illuminated by light that does not contain
that color at all, or that changes the received spectrum of light energy to
match exactly the spectrum of a different color seen under normal lighting
conditions. There is, in fact, no simple correspondence of perceived color
to the spectrum of incoming light.

We agree here, although I am choosing to emphasize another point: it is
possible to construct a correspondence between the frequency of light and
its perceived color when care is taken to avoid such problems as contrast
effects. Even so, that correspondence is not based on any ability inherent
in the retina sense the various frequencies of light which correspond to
those perceptions of color. It is constructed from the ratios of activation
of the three types of cones, which is the only color information passed from
the retina to the brain.

co-activation of these receptors and the near-absence of activity in the
highest-frequency receptor ultimately give rise to the perception we label
as yellow.

I don't think this view can be supported any more. There are too many
counterexamples, particularly those provided by Land.

Again, I was trying to keep the example simple in order to make the point:
the perceived color in both my examples is synthesized from the ratios of
activities of red, green, and blue photoreceptors (the cones). Under normal
conditions the output of the sodium lamp appears yellow, but there are no
yellow receptors in the retina, only receptors centered on other frequencies
that do not correspond to the frequency of sodium emission.

Isn't it much simpler just to say that yellow is the output of a particular
perceptual input function, without trying to claim that it represents
something real in the environment? Is there some reason NOT to say it that

It is simpler, but let us not overlook the fact that it may be possible to
determine how the characteristics of the input relate to the perception.
Saying that yellow is the output of a particular perceptual input function
tells us nothing about what the inputs of that function are. Although
perception of color of objects is influenced by the strength and coloration
of the light falling on the objects, by contrast with other colors, and
other factors, it is still the case that the colors of the spectrum visible
in light passed through the prism are there because of the prism's ability
to sort light waves out spacially by frequency. The perception of the
light's color changes in a systematic way with the frequency of light when
other factors are held constant and the light is in the right intensity range.

Similarly, variations in temperature seem to be conveyed as the intensity
signals of a single sense, but in fact the perception of temperature arises
from the co-activation of two, more specific, receptors in the skin. One
receptor becomes more active as skin temperature increases above some value
(the "warm" receptor). The other has an inverted U-function, becoming more
active as skin temperature falls below some value _or_ as the temperature
reaches an upper extreme. This latter receptor is labeled the "cold"
receptor and conveys the impression of "burning" hot when co-active with the
warm receptor. (Aside: I have omitted sensory adaptation effects in this
description, which complicate the picture somewhat.)

Fine, and that's interesting (I didn't know about the inverted-U-shaped
"cold" signal). But doesn't all this argue against the idea that there is a
simple correspondence between sensation perceptions and physical reality?

Yes. In fact, the same probably applies to what you call intensity signals
as well.



[From Bruce Nevin (980309.1414 EST)]

Bruce Abbott (980309.1400 EST)

It is constructed from the ratios of activation
of the three types of cones, which is the only color information passed from
the retina to the brain.

Not to pick on you, Bruce. I know, we gotta use words when we talk to one
another ... But "color information passed from the retina to the brain" was
too much of a stimulus for me not to respond ;->

Does a rate of firing of nerves leading from cones in the retina into the
brain constitute color information? Or is it just a light intensity
perception signal contributing (at a higher level of the hierarchy) to a
sensation perception that we call color?

Suppose it does constitute color information. If the same intensity
perception signal (rate of firing of the nerve) is on another occasion not
so combined with others as to contribute to a color sensation, is it no
longer color information?

(I know you said "ratios of activation". Contextual issues per Land's
findings bring us to the same question. In another visual context the same
ratio of signals from cone cells does not constitute the color.)

Calling it color information buys into the projection of a perception
(color) from the perceptual hierarchy into the environment. They ain't no
colors out there.

(I'm sitting here kind of amazed that this is not reductionism.)

My 16-year-old is a big Beatles fan. In "The Yellow Submarine" as the car
keeps changing color George says, "it's all in y'r mind, y'know." Well, of
course It isn't. Or They aren't. But all this stuff that the environment
appears to be is.

To the extent that the perceptual hierarchy is formed (in-formed) in the
course of controlling perceptual inputs, I suppose you could call the world
of perceptions "information" that is created by control. But the
information isn't in the environment whence it comes into the organism. The
information (if that is what it is) is constructed in the organism whence
it gets projected onto the environment.

  Bruce Nevin

[From Oded Maler (980310)]

Re: objectivity of color.

Just an anecdote, I visited last year a factory for industrial colors
and they showed me how their quality assurance works. You have pieces
of plastic with all the standard colors + catalogs saying what
a spectrometer should show for each of them. They take samples from
various parts of the color container and compare their spectra to it.
The objective agreement is needed here due to the contract between
the client and the manufacturer.