Greg Detre
Thursday, 18 May, 2000
colour = evolutionarily useful � depends on the environment where you find yourself
e.g. in the tropical reef environment, colour is probably the best way of identifying thing
the creatures there have up to hexachromatic colour vision (and our trichromatic vision is quite common), and they are usually vividly coloured
underwater, bad spatial resolution
colour is a good way of segregating � very difficult to distinguish colour + vertical simultaneously (serial???)
Kandel & Schwarz pathways segregation � outdated
there is separate processing, but there is some colour processing even in the magno-cellular pathway
now use fMRI machine to pick up changes in the oxygenation in the blood
the RGB cones do not actually correspond to those colours,
e.g. the orangey-yellow sodium street lights = very tight wavelength, but high up (560nm)
opponent theory vs trichromatic
subtractive (luminance channel)
parvo type 1 � most common
blue cones have their own special wiring (yellow-blue ganglion cell)
� Y-B layers in the interlaminar zones of the LGN, within the parvo region
cf the koniocellular system
blobs associated with specific colour signatures � perhaps
V4 = involved in many things, including colour
e.g. colour constancy
= if I have a yellowish bit of paper and measure the environmental light (reddish tungsten light in the lecture hall)
combination of absorption properties of the surface and the illumination