Re: Color Mixing (Pigment) question

[John Clement <clement@HAL-PC.ORG>]

Please, lets not make the texts even thicker.  Already they are so
overloaded with facts that students could never comprehend all of the
concepts.  They are a mile wide and an inch deep.  The best approach would
be to recognize that the texts are presenting a particular simplified model
which is of necessity incomplete because ultimately it involves some very
sophisticated biological processes which filter our perceptions.  In other
words we need to help the students understand that what they learn is always
subject to revision and elaboration.  Essentially all physics problems are
simplified in some way or other.  For example projectile motion problems
never take air resistance into account even though it does make a
difference.

The best approach is to have the students do various experiments and observe
the primary ideas.  It is easy to demonstrate light color mixing using a CRT
and also subtractive mixing using transparencies printed on an ink jet
printer.  And the students can also easily observe that the effects are not
perfect.

Essentially they are introduced to the engineering approach to color
reproduction.  However color reproduction of natural materials in various
lighting conditions does not obey the simple rules because the eye
essentially compensates for lighting conditions to keep the perceived colors
fairly constant independent of illumination.  It is possible to have two
color swatches reflecting exactly the same spectral distribution, but being
seen as entirely different colors.  This can be demonstrated by illuminating
color swatches on one side by a red spot, and by a blue one on the other. It
is possible to arrange the illumination so a red square on one side and a
green on the other have the same spectral balance, yet still appear with the
same color as viewed in white light. Some of this has been alluded to with
respect to the impressionist and pointillist artists in other posts.  The
Retinex theory by Edwin Land is an example of a model that attempts to take
into account the more complex features of color vision.

The engineering approach does work very well with TV and color
printers/photos.  For them the usual color wheel is abandoned in favor of
the chromaticity diagram.

This topic has been around before, but not in the context of a textbook
answer which is wrong in the context of the model that was presented to the
students.

If you want to supplement the texts for a course which has the time to spend
on a deep treatment of color, a good online book can be found at the Howard
Hughes Medical Institute.
http://www.hhmi.org/senses/

John M. Clement
Houston, TX

> Actually, I have done the experiment, but I admit that I have not done it
> recently. As per the ongoing discussion, the physics will differ
> considerably depending on whether you're using oil paints, water colors,
> printer dyes, crayons (remember crayons?), colored chalk, pastels, color
> filters, computer programs, or mixing colored powders such as those used
> in
> sand art. Some of the posts in the current thread are very interesting. It
> is clear that there is a lot of confusion and oversimplification in the
> texts. So even though color mixing is a physics topic in which
> experimentation is frequently relegated to elementary school art classes,
> the physics is far from simple.  Perhaps color is a topic that deserves
> far
> more textbook space than it usually gets.
>
> Vickie