National Geographic

The Mantis Shrimp Sees Like A Satellite

The most extraordinary eyes in the animal kingdom belong to the mantis shrimps, or stomatopods—pugilistic relatives of crabs and prawns, which are known for delivering extremely fast and powerful punches. Their eyes sit on stalks and move independently of one another. Each eye has “trinocular vision”—it can gauge depth and distance on its own by focusing on objects with three separate regions. They can see a special spiralling type of light called circularly polarised light that no other animal can. And they have a structure in their eyes that’s similar to technology found in CD and DVD players, only much more effective.

And now, Hanne Thoen from the University of Queensland has found that mantis shrimps see colour in a very different way to all other animals.

Most people have three types of light-detecting cells, or photoreceptors, in their retinas. These are sensitive to red, green and blue light, respectively. Birds, reptiles and many fish have a fourth photoreceptor that detects ultraviolet light. Four is plenty. Mathematical models tell us that you only need four receptors, maybe five, to effectively encode the colours within that range.

The mantis shrimp has twelve different photoreceptors.

Eight of these cover the parts of the spectrum that we can see, while four cover the ultraviolet region. That seems like a ludicrous excess. If four or five receptors are all an animal needs, “why on earth do stomatopods need 12 channels?” says Justin Marshall, who led the new study.

The obvious answer is that they’re very good at discriminating between different colours. That would be a handy skill: mantis shrimps live in coral reefs, which are bursting with colours. Many of them are brightly coloured themselves, and use their lurid body parts to communicate with one another. “With 12 receptors, you’d think that they can detect colours much better than any other animal,” says Marshall.

“Actually, they’re much worse!”

Thoen discovered their surprising ineptitude by studying a small species called Haptosquilla trispinosa. She presented the animals with two optic fibres, each displaying a different colour. If they attacked the right one, they earned a tasty snack. Thoen then changed the colour of the off-target fibre to the point when the mantis shrimp could no longer tell the difference between the two.

If a human did this test, we’d be able to tell the difference between colours whose wavelengths are 5 nanometres apart—compare the left and middle columns in the image below. A mantis shrimp would struggle with that. The can only discriminate between colours with a 15-25 nanometre difference—compare the middle and right columns.


A human could tell the difference between the colours in the left and middle columns with a 50% accuracy. A mantis shrimp could only do the same for the colours in the middle and right columns.

Despite their 12 photoreceptors, mantis shrimps are worse at telling apart different colours than humans, honeybees and butterflies.

“Thoen is a very careful scientist, so the data are completely convincing, if quite surprising,” says Tom Cronin from the University of Maryland, Baltimore County, who studies mantis shrimp vision. “We certainly would have predicted a much more competent sense of color discrimination than this!  However, behaviour is the ultimate test of what an animal can do, so this is what the animals say that they are capable of.“

They must be using the information from those receptors in a very strange way.

We see colours by making comparisons between our three receptors. By comparing the outputs of the red and green receptors, we can tell the difference between reds and greens. And by comparing their combined output against that of the blue receptors, we can discriminate between blues and yellows. This is called the “colour opponent process” and it’s what every colour-sighted animal does.

Every animal… except the mantis shrimps. Given their poor performance in Thoen’s tests, they cannot possibly be making these comparisons. What are they doing instead?

“The simple answer is: Dunno,” says Marshall. “I’ll admit right up front we don’t fully understand.”

Their working hypothesis is that the mantis shrimps analyse the outputs from all of their 12 receptors at once. Rather than making comparisons between those receptors, they pass the entire pattern of outputs onto the brain, without any processing. “One could imagine that they have a look-up table in their brain,” says Marshall. So rather than discriminating between colours like we do, their eyes are adapted for recognising colours.

“Oddly enough, the closest device to stomatopods would be a satellite,” says Marshall. “Remote sensing algorithms have look-up tables of colour to fill in the image that the satellite forms.”

Marshall suggests that this way of dealing with colour should be much faster than ours, since there is no need to send the photoreceptors’ signals through any intermediary neurons. And speed matters for mantis shrimps. These ambush hunters attack their prey with rapidly unfurling arms, which end in either stabbing spears or pounding clubs. The clubbed species, known as smashers, can hit their targets with the force of a rifle bullet and deliver the fastest punches in the animal kingdom. They need fast eyes to complement their fast arms.

And they only have a small brain. “A mantis shrimp only has a fraction of our cortical processing power, yet it handles 4 times more input,” says Nicholas Roberts from the University of Bristol. “The non-comparative processing system they have evolved represents a novel solution for increasing data acquisition while minimising any downstream processing overhead.”

Of course, this is still conjecture. Thoen and Marshall have shown that mantis shrimps definitely don’t see colours in the same way as us, but what they actually do is a mystery. Now, they’re trying to work out what happens to signals when they leave the photoreceptors, and how these cells are connected to the brain.

Cronin also wants to know “whether these animals combine their colour receptor signals in different ways for different tasks. Perhaps analysis of mate displays or colour signals demands a more thorough discrimination than food recognition.”

Marshall adds that the mystery is relevant to one of the most important questions in neuroscience: How does a nervous system make sense of information from the outside world. “This is clearly a very different way of computing that information,” he says. “It’s not just about weird shrimp biology. It touches on a number of neuroscience questions.”

Reference: Thoen, How, Choiu & Marshall. 2013. A Different Form of Color Vision in Mantis Shrimp. Science

More on mantis shrimps:

There are 6 Comments. Add Yours.

  1. Stephan Zielinski
    January 23, 2014

    There’s none so blind as those who will not see.

  2. Dale
    January 23, 2014

    I don’t think you meant to say that no other creature can see circularly polarized light. Surely, no other creature can DISCRIMINATE between circularly polarized light and other light, but they absolutely can see it; I know, because I watched a movie in 3D, which relies on my ability to see circularly polarized light.

  3. Craig
    January 30, 2014

    Dale, humans cannot detect circularly-polarized (actually spiral) *without the 3D glasses.
    The un-polarized movie images are encoded to circular-polarization at the projector, to be decoded to two separate images, one for each eye.

  4. Wang-Lo
    January 30, 2014

    @Craig: I can’t let your remark stand uncorrected in a venue where people may come to learn science. There is not “decoding” in the stereo eyeglasses, just filtering.

    As you say, the un-polarized movie images are encoded to circular-polarization at the projector. The viewing glasses have one lens polarized clockwise and the other polarized counterclockwise. Each lens blocks the image intended for the other eye.

    The light that passes through the lenses is circularly polarized — clockwise for one eye and counterclockwise for the other. The human eye has no difficulty responding to this light.

    If you view the screen without the glasses you will see the doubled image, with each of your eyes detecting both images. All the light you are seeing is polarized.

  5. Peter Ralph
    February 1, 2014

    I hate to just put in a correction, but this correction is a really cool fact: besides mantis shrimp, a number of beetles can see circularly polarized light — their carapaces reflect circularly polarized light, too, so presumably use this as a way to find each other (ref: ).

  6. Peter Ralph
    February 1, 2014

    Ooops; you said “no other animal” — so, strike “correction”, and that’s just a cool fact in my previous comment.

Add Your Comments

All fields required.

Related Posts