"New" colours possible?

Vincent Ferrera vpf3 at columbia.edu
Thu Jan 3 11:43:54 EST 2002


Urs Enke wrote:
> 
> For some years now I've wondered
> -- whether the (red-green-blue-mixable) colours we know are all there are in
> this universe, and if not,
> -- whether it might be possible to neurologically change the visual cortex
> (or whatever necessary) to create the sensation of other colours, and
> -- whether there has been any research suggesting that other animals are
> actually seeing different colours than we do.
> 
> I am neither talking of broadening the EM-spectrum we can see by changing
> the eyes' perception (as this would supposedly simply distribute our known
> "rainbow colors" over that new spectrum), nor of mixing pseudo-new colours
> from the base colours that we already know. Also, I wouldn't consider any
> eye-related limitations essential, as I really mean directly tapping into
> one's brain.
> 
> I'd be glad to get any feedback, be it on physical possibility, neurological
> practicability or philosophical thoughts. Thanks in advance for brightening
> or dimming my hopes to see "Color X" before I die... ;-)
> 
> Urs
> 
> PS: Maybe someone even has information on people reporting previously
> unknown ("unmixable") colours after having been under the influence of
> drugs...?

For a trichromat, every color can be represented as a three-dimensional
vector, or "tri-stimulus value", in some color space (RGB, HSV, etc.). 
So the question is whether the responses of "color" cells in the visual
cortex are also constrained to a three-dimensional space.  If there are
only three color cells for each retinal location, then it is possible
that there is a one-to-one mapping between tri-stimulus values and
patterns of neural activation such that there are no neural states that
do not correspond to tri-stimulus values.

However, if there is any redundancy in the neural representation, then
the possible patterns of activity might not be limited to three
dimensions.  For example, you could have two neurons with the same
spectral sensitivity and those neurons would always have the same level
of activity for any natural stimulus.  But you could, in principle, use
microelectrodes or some other artificial means to stimulate them so they
would fire at different levels.  This pattern of activity would be
different from anything that could be produced by a natural stimulus and
might correspond to a color the subject had never seen before, i.e.
color X.

The question is, if you succeeded in creating a unique pattern of
activity among the color cells in your visual cortex, a pattern that
didn't correspond to any natural stimulus, would you recognize the
resulting percept as a color?




More information about the Neur-sci mailing list