This post draws on a guest lecture, titled ‘Colours and Consciousness’, delivered to Lancaster University’s Philosophy Society in November 2022. The central questions asked there were:
Why do colours look the way they do?
Might you see green where I see red (and vice versa)?
If so, can we tell?
These questions are more philosophical than scientific, since it is hard to know even where to look for answers. But they were first asked by the ancient Greeks, and they still have an important role in the modern study of how the mind relates to the body (the ‘mind-body problem’).
We know a great deal about the physics and physiology of colour perception, of how light interacts with the electrons on the surfaces of objects, and of how the eye and brain interpret incoming data, but the actual phenomenology is not thereby addressed. Philosophers talk here of an explanatory gap: why do we see things as depicted in Figure 1 and not Figure 2?
Figure 1
Figure 2
The above photographs provide an example of a sensory inversion. Examples of other sensory inversions include:
Could you see things get darker where I see them get brighter?
Could you hear a soft sound where I hear a loud one?
Could you hear a high-pitched sound where I hear a low-pitched one?
Could you hear the sound (timbre) of a trumpet where I hear that of a clarinet?
Could you taste things bitter where I taste them sweet?
Could you feel a pleasurable sensation where I feel pain?
None of these inversions seem psychologically at all plausible. This suggests that there is something peculiar about colour hues that resists parallel explanation. A way forward, perhaps, is to focus on what this is.
Colour-hues and physics
The visible wavelengths of the electromagnetic spectrum are located (roughly) between 400nm (violet) and 700nm (red). The perceived hue of an object depends primarily on the dominant wavelength of the reflected light. (But most purples are non-spectral, and can only be produced by mixture.)
The CIE 1931 colour space chromaticity diagram
The Hering Colour circle
Edward Hering was a psychologist and founder of modern colour vison science. There are four ‘unique hues’ - red, blue, green and yellow. These hues look essentially unmixed. There are also four ‘binary hues’ – orange (reddish yellow), purple (bluish red), turquoise (greenish blue), and chartreuse (yellowish green). These hues look essentially mixed.
Opponent processing
There are three types of cone in the eye, but two chromatic retino-cortical channels: the red-green channel and the yellow-blue channel (plus the achromatic light-dark channel). When the red-green channel is excited a reddish hue is perceived; when inhibited a greenish hue is perceived. (Likewise with yellow and blue.)
The physiological fact that the channel cannot simultaneously be excited and inhibited (any more than an energy level can simultaneously increase and decrease) has a phenomenological counterpart in the fact that no hue can be perceived as both reddish and greenish simultaneously.
Likewise, the physiological fact that perception of the unique (binary) hues requires activation of just one (two) channels has a phenomenological counterpart in the fact that unique hues look unmixed and the binary hues look mixed. This shows that physiology and phenomenology are more closely related than is sometimes thought.
Colour-hues and the eye
Warm versus cool hues
Some hues (red, orange, yellow) seem intrinsically warm (“advancing”, “positive”); other hues (blue, turquoise, green) seem intrinsically cool (“receding”, “negative”). There are cultural and physical associations here, but also (it seems) pure phenomenology. Perception of red/yellow involves excitation of neural pathways; of green/blue, inhibition. There is some (controversial) evidence that this excitation and inhibition has a wider physiological impact, which could perhaps explain these phenomenological features. If it can be shown that there really are these physiological-phenomenological connections, it would perhaps help us to explain why colours look the way they do. Maybe, the explanatory gap can be traversed after all. However, more work needs to be done. In particular, we need to address other possible colour inversions besides simple red-green inversion.
Hue-inversions
Simple red-green inversion involves reflection in the vertical axis. Red and green are interchanged: yellow and blue stay fixed. Diagonal inversion involves reflection in the dotted axis. Red is exchanged with yellow, and green with blue: orange and turquoise stay fixed. Diagonal inversion is less easily detectable than simple red-green inversion, since warm colours stay warm and cool colours stay cool.
Supersaturated Yellow
Yellow is lighter and has less intrinsic saturation (or ‘chromatic content’) than red, so in order to maintain undetectability we must suppose that diagonal inverts can perceive supersaturated yellow, a colour which relates to yellow as red relates to pink. We cannot perceive this. They (diagonal inverts), likewise, cannot perceive ‘supersaturated pink’ (i.e. red). Perhaps this is possible. If so, we need to take diagonal inversion seriously. The following slides depict photographs with such inversion.
A sunset with red-green and diagonal inversions
Holy Island, Northumbria; with red-green and diagonal inversions.
Rivington reservoir, Lancashire; with inversions.
Nick Unwin
Philosophy Lecturer at Lancaster University
Comments