Talk:Color vision

"Color vision table" -- Types of cones or types of color receptors?
The individual articles for monochromacy, dichromacy, etc. describe x-chromacy as having x types of color receptors or x independent channels for conveying color information. The "Color vision table" at the end of Section 2.5 In other animal species erroneously equates this with number of cones.

The existence of non-cone color receptors is well-described in the animal kingdom. For instance, many frogs and salamanders have dual-rod retinae wherein two different rod types are sensitive to different spectra (for a review of this, see e.g. this review, section "Amphibian Opsins and Photoreceptors"). This allows for color vision at the absolute visual threshold, i.e. scotopic conditions, or very low-light conditions. Animals have been shown to utilise this system under ethologically relevant conditions, as demonstrated by these experiments.

Species with two distinct color-sensitive rod types and two cone types are indeed tetrachromatic, yet they only have two types of cone cells, unlike what the table in this article would lead readers to believe. Additionally, the number of colors perceived is likely closer to 40,000 than 100 million due to significant overlap in the spectral sensitivities of their rods and cones.

To resolve these issues, I would recommend amending the table to entirely remove the column titled "State". Tetrachromacy does not mean four cone types and does not imply the ability to see 100 million colors, as we have seen from many studies of different species of amphibians.

NeuroJasper (talk) 14:55, 22 July 2020 (UTC)


 * After nine months with no further discussion, I have removed the Color Vision Table.
 * NeuroJasper (talk) 14:48, 21 April 2021 (UTC)

Neither article suggests that any of the animals are tetrachromats. Indeed, the second article explicitly states that the presence of photoreceptors with different spectral sensitivities that are functional at the same light levels is a requirement for colour vision. It then states that some some frogs may have dichromatic colour vision based only on their retinal rods. It does not state that they are tetrachromats, because they are not. You created an "issue" with the table by redefining "tetrachromacy" based on an incomplete understanding of how colours are discriminated using simultaneous signals from different receptors stimulated by the same wavelength. This is underscored by your further assertion that such tetrachromacy would be limited to the perception of only 40000 colours due to overlap in the spectral sensitivities of the rods and cones. All that is required is a different response between two types of receptors at any and all wavelengths to which they are sensitive. The range that they respond to can be exactly the same, and indeed, the long and medium cones in our eyes do have a very similar range, and very similar wavelength of maximum sensitivity. It is this that gives us our greatest colour discrimination in the orange to green part of the spectrum. In fact, if the rods and cones did work together at any light levels, the spectral overlap would increase the number of distinguishable colours. I believe you are confusing the number of distinguishable colours with total spectral range of vision. Distant spacing of spectral sensitivities increases the spectral range of colour vision, but decreases the ability to distinguish different wavelengths, requiring a larger difference in the wavelengths of two lights in order to perceive them as different colours. This is why humans can perceive a different colour with a change of 1nm in the spectral region covered by all three cones, but require a difference of about 10nm in those parts of the spectrum in which only two cones are sensitive. In the very long wavelengths, from 750nm to 950nm, considered infrared but visible with very bright, very pure sources, such as infrared LEDs, increasing the wavelength is perceived only as a decrease in brightness. The very deep red colour perceived does not vary, and it is impossible to determine if the "colour" has changed or if it has simply been dimmed. A sufficiently bright 900nm light will appear exactly the same as a dimmer 800nm light. The table was useful, and it was correct. You are in no position to estimate the number of different colours distinguishable with any number of different receptors, and until you find a published reference to the contrary, the chromacy of vision is determined by the number of different photoreceptors functional at a given level of light. By convention, this is based on cones, functional at photopic light levels. I will reinstate the table in a month or so unless you respond with some evidence that your assertions are in fact accepted by visual scientists. — Preceding unsigned comment added by At least I try (talk • contribs) 17:45, 19 December 2021 (UTC)