How mantis shrimp see circularly polarized light

Mantis shrimp have long been regarded as visual super-stars. They can have up to 16 distinct photoreceptor types that are maximally sensitive to at least 12 different wavelengths (colors) of light; from deep in the ultraviolet, across our visual range, and into the infrared. In addition, they are strongly sensitive to linearly polarized light (LPL) and are able to discriminate the angle at which these waves of light are oscillating as they travel through space. This visual modality, though hugely foreign to us, is actually well perceived by cephalopods, some chordates, and almost universally amongst the arthropods (I previously described linearly polarized light (LPL) and the way in which arthropods detect it here). Mantis shrimp however, seemingly never content to be matched, have taken polarization sensitivity a step beyond any other animal. They have evolved an ingenious means of detecting and discriminating circularly polarized light (CPL).

CPL is very rare in nature. Photons produced by stars are linearly polarized and oscillate in a flat plane as they travel through space as a wave. CPL is only produced when a LPL wave refracts through specific materials, throwing its oscillation out of phase, and causing the light wave to essentially corkscrew through space. CPL can be left-handed (L-CPL) or right-handed (R-CPL), depending on weather it is corkscrewing clockwise or counter clockwise.

A quarter-wave retarder converts LPL into CPL.

Because of its rarity in nature, and the inherent difficulty of detecting CPL, it was long thought that CPL was largely ignored by animal visual systems. However, when some visual modality seems unlikely, its always good to check with the mantis shrimp before completely ruling it out. And sure enough, certain mantis shrimp species posses specialized visual signaling structures that reflect CPL. They likely use these structures for covert intraspecific communication, undetectable to predators.

White L-CPL producing structure on the tail of the mantis shrimp Odontodactylus cultrifer, viewed through left- and right-handed CPL filters. From Chiou et al., 2008 and Chrissy Huffard.

Having observed these obvious CPL signaling structures on the bodies of mantis shrimp, researchers then set out to figure out if the shrimp were actually able to detect and discriminate CPL signals. They trained the mantis shrimp to attack feeding tubes with targets on them that reflected either L-CPL or R-CPL. They then removed the food reward and found that the shrimp still chose the tube they had been trained on at a rate significantly above chance. This clearly demonstrated that mantis shrimp could discriminate CPL, but how they did it still needed to be deciphered.

Specialized cells in the mantis shrimp retina act as quarter-wave retarders and convert CPL into LPL at opposite angles, depending on the handedness of the incoming CPL. Adapted from Chiou et al., 2008.

It turns out that no known photoreceptor can discriminate CPL on its own. However, remember that LPL can be converted to CPL by passing through a quarter-wave retarder. Well, that works both ways, and a quarter wave retarder can also convert CPL into LPL; which arthropod photoreceptors are excellent at detecting. And sure enough, in the mantis shrimp retina there are specialized cells sitting atop their largest LPL detecting photoreceptors, directly in the optical path. These cells are nearly perfect quarter-wave retarders, surpassing even the best synthetic retarders made by humans. They take incoming CPL, and depending on the direction it is rotating, convert the CPL into opposite angles of LPL.

This research makes the mantis shrimp the only animal that has been conclusively shown to detect and discriminate CPL (There may be a new contender however, more on this next). Mantis shrimp likely use this hugely uncommon visual modality to optically converse with their compatriots, while minimizing detection by visual predators. Their mechanism for detecting CPL is very elegant and evolutionarily interesting. This is especially true since the quarter-wave retarder cell is also a distinct photoreceptor itself; likely involved in detecting ultraviolet LPL. One wonders which function (retarder or ultraviolet LPL sensor) evolved first for these cells, since their morphology is integral for both mechanisms. Could they have arisen in tandem?

Whew… That was a doozy of a post for me to write. I always find the things I understand the best are also the most difficult to squeeze down into brief and understandable prose. Polarization vision especially makes me feel like a verbal contortionist as I painfully try to malleate it into shape for popular writing. Let me know if this article makes sense to you.


  • Chiou TH, Kleinlogel S, Cronin T, Caldwell R, Loeffler B, Siddiqi A, Goldizen A, & Marshall J (2008). Circular polarization vision in a stomatopod crustacean. Current biology : CB, 18 (6), 429-34 PMID: 18356053
  • Roberts, N., Chiou, T., Marshall, N., & Cronin, T. (2009). A biological quarter-wave retarder with excellent achromaticity in the visible wavelength region Nature Photonics, 3 (11), 641-644 DOI: 10.1038/nphoton.2009.189


  1. Mike Lisieski August 17, 2010 4:30 am

    I thought it was easy enough to read. I have read a lot on visual systems, though, so I might not be the best yardstick.

  2. j edward ladenburger August 17, 2010 10:23 am

    research always leads to additional questions… such as — are there results/data for response to elliptically polarized light?

  3. Michael Bok August 17, 2010 10:34 am

    I believe that question is being worked on.

  4. Nicholas Friedman August 19, 2010 10:48 am

    I was chatting with someone at the ABS meeting a few weeks ago who didn’t think these qualified as private signals. Their argument was that because these ornaments rely on underlying pigmentation that is conspicuous without the aid of CPL discrimination, they’re not really that private. Does this sound accurate? I don’t know much about the system, including whether those colors in Short’s figure are from pigments or structural elements (or both).

    • Michael Bok August 19, 2010 10:59 am

      I think the point is that, while other animals can see the light that is reflecting off of the telson’s keel, to them it is just a broad-band white reflection. Other mantis shrimp are the only animals that can detect that it is this very particular flavor of circularly polarized white light. It is going to stick out a lot sharper for them since their visual system is actually able to look for this stuff.

      To a predator, the keel may be a white rock, or a piece of coral, but to another mantis shrimp it is the keel of a buddy (or rival, or potential mate); there are very few other strong sources of CPL in the environment.

      • Nicholas Friedman August 20, 2010 3:58 pm

        Thanks Mike, that clears it up a lot for me.

    • Mike Lisieski August 21, 2010 12:16 am

      This can also be considered a private signal because, even though the pigmentation is conspicuous without a polarization filter, there are patterns in it that can only be discriminated when the observer can perceive circularly polarized light (at least, this appears to be the case in the photos above, where two parts of the tail appear to have opposite circular polarizations.) While the object can be perceived with “normal” vision, it lacks some of the information that is there when viewed with “polarized” vision.

      I don’t know how much information those tail patterns carry, though, besides simply being more noticeable to other mantis shrimp. @ Mike Bok – has anybody studied whether the patterns of circularly polarized reflected light differ between mantis shrimp of different ages/attitudes/sexes/genetic pools/nutritional statuses, etc.? If they do, the patterns might carry information to conspecifics that is invisible to other animals.

      • Michael Bok August 22, 2010 7:46 pm

        The CPL reflecting structures are found only on males so far.

  5. Mike from Ottawa August 20, 2010 8:48 pm

    That’s the first time circular polarized light has made even vague sense to me. Thanks.

    Mantis shrimp. Is there anything they can’t do.

  6. Whysharksmatter August 23, 2010 5:26 pm

    Just so we’re clear, these guys still can’t open doors, right? We’re safe as long as we have locks?