Paint me like one of your mantis shrimps

Sunsets, wildflower superblooms, unicorn frappuccinos. None of these phenomena would have nearly the same allure if we lost our ability to see different colors. But what if we could have an even greater ability to discriminate between colors – would these phenomena be that much more spectacular?

In the case of the frappuccino, probably not, but for the others, it’s enticing to think about. Are there those who can see more colors than the average person? Do the incredible eyes of the mantis shrimp make its world colorful to a degree that we can’t possibly imagine? To answer these questions, we must consider what in our eyes – and in our brains – allows us to perceive a colorful world. It’s a little more complicated than you might think…


A-B-C, 1-2-3, L-M-S…

At the most basic level, our color vision is enabled by our unique collection of photoreceptors – neurons in the retina of the eye that absorb light and transform it into electrical signals. Photoreceptors come in two major varieties: rods, which are specialized for vision in low-light conditions (like night time), and cones, which enable our visual acuity and – of primary interest for this article – color vision.

While there is only one type of rod, cones come in three different varieties, each sensitive to particular wavelengths of light, i.e., color. Though they are sometimes referred to as “red”, “green”, and “blue” cones, a more appropriate naming scheme is by the length of wavelength that each cone type absorbs best: long (“L”), medium (“M”), and short (“S”) (see Figure 1). Together, this trifecta of cones, each with a slightly different preference for wavelengths of light, gives rise to our perception of color (not unlike how computer and TV screens display multicolor images using varying amounts of only red, green, and blue).


Figure 1: Spectral tuning curves of S, M and L cones, depicting how well each cone type absorbs different wavelengths of light (image from David Briggs 2014)

Imagine it’s Fall and you are admiring a beautiful apple tree. Some apples are red and ripe for picking, while others are still green and have not yet reached full maturity. Each apple highlighted in Figure 2 reflects a different wavelength of light, which is absorbed to different degrees by each type of cone. This leads to distinct activation patterns in your retina that get passed along through the rest of the visual pathway in the brain, ultimately giving you your perception of one apple as “green” and the other as “red”.


Figure 2: A (hypothetical) schematic depiction of how well each of the different cone types absorbs light reflected off each of this different apples.

Now imagine what would happen if you were missing one or more of your three types of cones. With only a single type of cone, you wouldn’t be able to make any distinction between light’s wavelength (hue) and its intensity (brightness). For instance, if you were left with only the hypothetical M cones depicted above, they could hypothetically be equally activated by the green apple and a red apple that is brighter (but still red) that the one depicted above. Thus, your brain would have no way of “reading out” the apples’ distinct colors independent of their brightnesses. In this case, you would be colorblind [1].

However, if you have ever known someone who was “colorblind”, they most likely weren’t truly colorblind but rather “blind” to the difference between certain colors. This is the result of dichromacy, i.e., missing a single type of cone (leaving the other two intact). For most dichromats, either the M or the L cones are delinquent, which causes them to have considerable trouble distinguishing between red and green [2]. In the prior hypothetical example, since the M and L cones were primarily responsible for discriminating between the red and green apples, the loss of one of those cone types would result is a similar pattern of photoreceptor activation for both apples and ultimately your perception that they are the same color. This is essentially the principle behind the most common color vision test used to assess color deficiency (demonstrated in Figure 3).


Figure 3: A typical test of red-green color deficiency. A) The normal image, with a number embedded inside in a different color. Can you see it?! B) What A) would look like to people missing their L cones. C) What A) would look like to people missing their M cones. D) What A) might look like to people with a hybrid L/M cone (anomalous trichromacy).


But why are the M and L cones more prone to getting messed up than the S cones? The genes that encode the light-absorbing pigments that make them either “M”- or “L”-type are very similar to each other and also closely located on the X chromosome, thus rendering them particularly susceptible to genetic mistakes. This also explains why red-green colorblindness is significantly more common in men than in women: approximately 8% of men and less than 1% of women [1]. Since women have two X chromosomes, if one of them is missing the L or M genes, they’ve still got back-up copies on the other chromosome, resulting in normal L and M in addition to S cone pigments. But for men (XY) who only have one X chromosome, it’s one strike and their red-green color discrimination is out.

Not only does a woman’s second X chromosome protect her from red-green colorblindness, it also has the remarkable potential to add an additional cone type. In some cases (referred to as “anomalous trichromacy”), one of the pigment genes on the X chromosome might get a little contaminated by the other gene, resulting in one regular M or L gene and another “hybrid” L/M gene [1]. Men with this genetic abnormality have reduced sensitivity to red-greens, but may still be able to recognize the number hidden in the image above (Figure 3D). But for a woman, if one chromosome ends up with an L/M hybrid gene as discussed above, but the other chromosome has the normal L and M genes, she would end up with FOUR cone types: S, M, L, and an L/M cone with a wavelength sensitivity somewhere between that of the L and M cones [1]. While this is known to be the case genetically, evidence that these women are behaviorally tetrachromatic – in other words, whether their visual systems actually make use of the other cone type to provide an extra degree of color vision – is surprisingly limited. For instance, a 2010 study found that only one out of 24 women they tested with this genetic idiosyncrasy demonstrated advanced color discriminability, such as being able to tell the difference between “yellow” that was a mix of red and green wavelengths and a “pure yellow” [3]. Still, the fact that even one of these women demonstrated a degree of extra-ordinary color vision abilities is fascinating. How might she perceive the world differently than the rest of us do?!  

If it isn’t already mind-boggling enough to consider what it would be like to have an additional cone photoreceptor and be behaviorally tetrachromatic, consider some of the color freaks of the animal kingdom. Birds have five types of cones [4], butterflies have at least that many [5], and – most famously – the mantis shrimp has at least TWELVE! But does this necessarily mean that their color vision is that much more sophisticated than ours? Remember, for us and our fellow primates, color perception doesn’t arise in the eye itself; the eye must send the color information to the brain where it undergoes many transformations before we ultimately perceive color.


Color is in more than the eye of the beholder

After leaving the eye, signals from the three different cone channels begin to converge in the brain, and this is where things really start to get interesting. Neurons in the next stop of the primate visual pathway after the eye – the visual part of the thalamus – may increase or decrease in activity, depending on the color of the stimulus (a principle called color opponency). From there, signals get sent to the primary visual cortex, and color contrast begins to emerge: cells may increase their activity to a red spot of light with a green border, but decrease their activity to a green spot of light with a red border!

As these color-coding signals continue to move through the brain, they become increasingly complex so that by the time you get to another part of cortex – the posterior inferior temporal cortex (PIT) – neurons change their activity in response to specific colors beyond just red, blue, green and yellow. If a neuron is “purple-tuned”, it will increase its activity in response to purple stimuli but not to stimuli of any other colors, and its activity will increase regardless of whether the purple stimulus is very dim or very bright [6]. And while some populations of PIT neurons are “tuned” to purple wavelengths, others are tuned to pink, or orange, or red, or green, etc as depicted in Figure 4. Truly, these are “color neurons”, and we can start to see (pun intended) how our color perception emerges.



Figure 4: Tuning curves of individual neurons in PIT cortex of a macaque monkey (which have extremely similar visual systems to our own!) [7].

Do you still think that the mantis shrimp’s 12 photoreceptors would give it color vision that is lightyears beyond our own? Compare the PIT tuning curves to the spectral tuning curves of photoreceptors in the mantis shrimp:




Figure 5: Spectral tuning curves of individual types of photoreceptors (left; [7]) in the retina of the mantis shrimp (right). Compare this to the spectral tuning curves of individual PIT cortex neurons (above) and of humans’ three cone photoreceptors from Figure 1!

While the huge range of wavelength sensitivity in the mantis shrimps’ photoreceptors is certainly impressive, it doesn’t look all that different from the range of wavelength sensitivities of neurons in primate PIT cortex. It may be that, with the exception of having three whole photoreceptors dedicated to seeing ultraviolet (which is undeniably awesome), the mantis shrimp’s color vision may not actually be any more sophisticated than ours. What the mantis shrimp does in its eye, we do in our brains [7]; though we only have three channels of color in our eyes, our visual systems directly compare activation across these channels in order to see many more than just three colors. If the mantis shrimp’s brain utilized cross-channel comparisons like our brains do, their color discrimination ability would be predicted to be much, much better than it actually is (and yes, scientists have actually trained mantis shrimps to play a color discrimination game) [8].

So, there’s no need to be too envious of the mantis shrimp’s vision (although they have plenty of other superpowers of which to be envious, like using exploding bubbles to deliver sonic punches to destroy their enemies). Our color vision is the complex product of the photoreceptors in our retinas, their associated pathways in our brains, and the neurons in different visual compartments of the brain that are specialized for making complex computations on those color signals. The result is an incredibly rich and multidimensional perceptual experience of the colorful world around us. So the next time you are admiring a sunset, don’t think about how much better and more beautiful it would be if only you were a mantis shrimp; think about how, just maybe, the mantis shrimp should be envious of you.



Featured image: modified from prilfish

Spectral peaks (Figure 1): David Briggs 2014

Color vision testing (Figure 3) :

Mantis shrimp (Figure 5):



[1] Conway, B. R. (2009). Color Vision, Cones, and Color-Coding in the Cortex. The Neuroscientist, 15(3), 274–290.


[3] Jordan, G., Deeb, S. S., Bosten, J. M., & Mollon, J. D. (2010). The dimensionality of color vision in carriers of anomalous trichromacy. Journal of Vision, 10(8), 12–12.

[4] Kram YA, Mantey S, Corbo JC (2010) Avian Cone Photoreceptors Tile the Retina as Five Independent, Self-Organizing Mosaics. PLoS ONE 5(2): e8992.

[5] Arikawa, K. (2017), The eyes and vision of butterflies. J Physiol, 595: 5457–5464. doi:10.1113/JP273917

[6] Conway, B. R., Moeller, S., & Tsao, D. Y. (2007). Specialized color modules in macaque extrastriate cortex. Neuron, 56(3), 560–573. doi:10.1016/j.neuron.2007.10.008

[7] Zaidi, Q., Marshall, J., Thoen, H., & Conway, B. R. (2014). Evolution of Neural Computations: Mantis Shrimp and Human Color Decoding. I-Perception, 5(6), 492–496.

[8] Thoen, H. H., How, M. J., Chiou, T.-H., & Marshall, J. (2014). A Different Form of Color Vision in Mantis Shrimp. Science, 343(6169), 411.