Pharyngula

The superior eyes of shrimp

Blogging on Peer-Reviewed Research

i-77f94af0d8588e45a64631732d5853b3-shrimp_eyes.jpg

We mammals have been beaten again. Shrimp have more sophisticated eyes than we do, with the ability to see things we can’t, and I’m feeling a bit envious.

There are a couple of general properties of light that can be captured and measured with a light detector. One is the amplitude of the light wave, which we see as differences in the intensity of light. This is the most basic measurement of a photoreceptor, sensing the raw amount of energy being transmitted. Another property is wavelength, which we perceive as the color of light. Many mammals are incapable of detecting the wavelength, having monochromatic vision, so we’re actually among the fortunate lineages that have a sufficiently elaborate photoreceptor system that we can actually see the color of objects.

Don’t get cocky, though: there are some features of light that we can’t see, but that would be very useful. One is phase. You can have two light waves of equal amplitude and wavelength, which look identical to our eyes, but they can be a fraction of a wavelength out of phase relative to one another.

i-1b32d3cea946e0272c5010d185d1f5d8-phase.jpg

That may not sound particularly useful to you because you’ve never experienced it. However, many materials shift the phase of light as it passes through them; a transparent (to our eyes) piece of glass can actually change the phase of light, as can cell membranes. We can build instruments to detect phase shifts, and they are extremely useful in microscopy — by aligning light waves that pass through a specimen with an unperturbed reference wave, we can generate patterns of constructive and destructive interference that our eyes can see, and voila, the transparent cell is made visible.

One other ability that we lack, but that many other animals possess, is light polarization detectors. Again, this is a property you may not be able to appreciate until you experience it, but it’s another powerful visualization tool. Light waves can be vibrating in any of multiple planes. For instance, the plane of vibration can be up and down, or it can be side to side, or at any angle in between. Light can also be circularly polarized, where the plane of vibration corkscrews through space. (How, you may wonder? It’s actually not hard; circularly polarized light can be decomposed into two linearly polarized light waves of equal frequency, but with the planes orthogonal to one another and with each out of phase with respect to the other).

We can’t see the polarity of light! Up and down, side to side, it’s all the same to our eyes, which is unfortunate, because we are often in environments where the light is polarized. Sunlight, for instance, is partially polarized as it passes through the atmosphere, and has a plane of polarization perpendicular to the direction of the sun, a property that many insects can use for navigation. Light reflected off of shiny surfaces is also selectively polarized. Polarizing sun glasses, for instance, are useful because than can selectively filter out glare reflected from horizontal surfaces. As light passes through otherwise transparent materials, like, say, a transparent animal swimming through the ocean, it can also be selectively polarized, a useful property to take advantage of if you are hunting for small, nearly transparent animals to eat.

The cephalopod eye has polarization detectors, but if you really want some sophisticated light detection, look to the arthropods, especially mantis shrimp. Even a superficial view of the eye reveals that something special is going on: there are specialized ommatida (the facets of the compound eye) and a curious equatorial band of aligned and specially organized receptors.

i-9f71a4a4fbac5f40f572316ba2d767c0-polar_eyes.jpg
A) Adult Gonodactylus smithii, or mantis shrimp, ˜7 cm long. The stalked apposition compound eyes are divided into a dorsal and a ventral hemisphere by an equatorial mid-band of enlarged and structurally specialised ommatidia. Inset. Mid-band position indicated by curved dark lines. The three pseudopupils (dark spots) visible within each eye indicate that the visual fields of the two hemispheres and the mid-band almost completely overlap at the equator of the eye, so that the three eye regions view the equatorial strip simultaneously. Photograph by R.L. Caldwell. B) Frontal view of the right eye to illustrate the division of the eye into a dorsal hemisphere (DH) and a ventral hemisphere (VH) by the equatorial mid-band formed by six rows of enlarged ommatidia, numbered row 1 to row 6 from dorsal to ventral. Mid-band rows 1-4 contain spectral photoreceptors; mid-band rows 5 and 6 are specialised for circular polarisation vision; the dorsal and ventral hemispheres for linear polarisation vision. Recording electrodes were lowered through corneal holes cut in the lateral half of the dorsal hemisphere, where the mid-band is ˜15° relative to the equator of the eye. The black scale bar is 1 mm, the axes refer to Dorsal, Medial, Ventral and Lateral. C) Electron micrograph of a longitudinal section through a mid-band row 6 rhabdom. The alternating layers of microvilli are highly ordered and in thinner layers than in hemispheric rhabdoms. The polarisation discrimination D of mid-band rows 5 and 6 retinular cells is twice as high as that of hemispheric cells due to a more crystalline microvillar structure: c.f. D̅mid = 0.340±0.061 with D̅hemi = 0.145±0.035

Kleinlogel and White dug deeper into that eye and found that the structures specifically support detection of polarized light, both linearly and circularly. They inserted microelectrodes into the photoreceptors and measure the responses as the eyes were flashed with polarized light, and got some beautiful electrophysiological data that showed wonderfully clean response curves. They have optimal polarization vision, capable of resolving all of the parameters necessary to detect the plane of circularly and linearly polarized light in their field of vision. This is powerful stuff; mantis shrimps are moving through a visual world rich with novel details beyond our imagination, able to detect qualities of light outside our experience. The authors see the importance, too.

Stomatopods are shallow-water crustaceans in a visual environment with a partially polarised background. Crustaceans are known to use polarisation for navigation; many stomatopod prey species are either reflective or transparent but change the polarisation of the light—an obvious possible driver of evolutionary change. Optimal polarisation vision provides all the information about polarisation of the visual field without confusion states or neutral points—giving the greatest ability to detect changes in both the degree and type of polarisation. This goes beyond simple contrast enhancement: optimal polarisation vision is analogous to the improvement afforded by stereo over mono vision in terms of increased information capacity.

The only conclusion possible is that it is completely false that God hates shrimp — he must like them very much indeed, and the biblical injunction against eating them must be because he likes them better than us.


Kleinlogel S, White AG (2008) The Secret World of Shrimps: Polarisation Vision at Its Best. PLoS ONE 3(5): e2190. doi:10.1371/journal.pone.0002190.

Comments

  1. #1 alex
    May 24, 2008

    that’s pretty amazing. i’ve got a feeling that this won’t be the last time i strongly envy shrimp.

  2. #2 Coturnix
    May 24, 2008

    You can send a trackback to the paper if you want.

  3. #3 xavi
    May 24, 2008

    With some practice humans too can detect polarized light

    http://www.polarization.com/haidinger/haidinger.html

  4. #4 andy
    May 24, 2008

    Well, in a way it’s a good thing we can’t see these properties of light: think how much more difficult life would be for artists, or people building TV/computer screens!

    (sorry if this comment appears multiple times – it appears ScienceBlogs and NoScript don’t get along…)

  5. #5 MIkeG
    May 24, 2008

    Now that’s just cool stuff.

    Thanks, PZ!

  6. #6 TomS
    May 24, 2008

    My understanding is that “mantis shrimp” are not shrimp. (I’m just going by Wikipedia on this.)

  7. #7 George
    May 24, 2008

    PZ – Plug your Bloggingheads diavlog with John Wilkins –
    http://bloggingheads.tv/diavlogs/11305

    Admittedly, poor quality but a great conversation.

  8. #8 scooter
    May 24, 2008

    This thread is making me hungry

  9. #9 Matt Heath
    May 24, 2008

    Argh. being a chordate blows

  10. #10 David Marjanovi?, OM
    May 24, 2008

    With some practice humans too can detect polarized light

    That is nothing, and it doesn’t even work for everyone!

    Many mammals are incapable of detecting the wavelength, having monochromatic vision

    Isn’t bichromatic vision — the ability to distinguish blue and yellow (red-green-blindness) or red and ultraviolet — much more common?

    ommatida

    Ommatidia. The figure legend has it right.

  11. #11 The Science Pundit
    May 24, 2008

    This thread is making me hungry

    Me too!

    PZ – Plug your Bloggingheads diavlog with John Wilkins -
    http://bloggingheads.tv/diavlogs/11305

    Admittedly, poor quality but a great conversation.

    Ooh! Watching it now.

  12. #12 wazza
    May 24, 2008

    I understand they’re fairly intelligent, too.

    You know, for crustaceans.

  13. #13 C Barr
    May 24, 2008

    Wow, so now does that mean we have to re-order the Great Chain Of Being?

  14. #14 Mike Haubrich, FCD
    May 24, 2008

    The development of vision pathways is such a great example of showing how evolution works to develop traits. I only wish that our common ancestors with the mantis shrimp had developed this before the great branching off.

    Now, if we could find some developmental biology research to show how their development differs from ours in larval/embryological stages of eyes…

  15. #15 JM Inc.
    May 24, 2008

    Hey, I’d still take being human over being an arthropod. Give me the big brains. I say we invent some eyes that can do all that – we’re already plugging totally blind people into cameras to give them some amount of vision (it’s just really, really poor vision; as it turns out, sight is harder to engineer than hearing). Some day we’ll all have this sort of vision too – if there’s anything humans are, it’s niche-fillers; we figure out what to do and beat everybody else at their own game.

    Cool stuff, though, PZ! Next week: Nematodes!

  16. #16 Stephen Bahl
    May 24, 2008

    I, for one, am ready to serve our stomatopod masters when they take to the land. I have no desire to be smashed or speared by their claws, nor trampled beneath their many feet.

  17. #17 Karen
    May 24, 2008

    We don’t have to re-order the Great Chain of Being, just order the Shrimp Scampi, Prawns in Lemon Sauce, or Shrimp Tempura.

  18. #18 C Barr
    May 24, 2008

    That spread would make even my eyes bug out.

  19. #19 MIkeG
    May 24, 2008

    We don’t have to re-order the Great Chain of Being, just order the Shrimp Scampi, Prawns in Lemon Sauce, or Shrimp Tempura

    And Gorton’s Law is fulfilled.

    Now, pass the melted butter!

  20. #20 Wilson Fowlie
    May 24, 2008

    #15: No, not nematodes; coral-dwelling gobies! (See below)

    Bob MacDonald of the Quirks & Quarks science show interviewed Dr. Kleinlogel this past Saturday (May 17). There’s a downloadable MP3 of the interview at the link.

    Her interview is the third item on the page, but the others are interesting, too. I’d be interested to hear if PZ has comments about coral-dwelling gobies deliberately restricting their feed intake to be 93% or less of the size of the next-biggest hierarchy in the group.

    I have to say, I didn’t like the way that interviewee – Dr. Marilyn Wong of McMaster University – characterized (almost anthropomorphized) the mechanism behind that behaviour, at the end of the interview. But the science was still pretty cool.

  21. #21 Wilson Fowlie
    May 24, 2008

    Drat. I should proofread better. The food intake isn’t 93% of the size of the next-biggest member of the group; each member restricts its food intake so that she* is the required proportion smaller than the next-biggest fish in the hierarchy.

    *All of the fish in each group are female, except one. If I recall the succession correctly, the 2nd-biggest member of the group becomes male for the period that he is in that position; when the biggest fish (a female) dies off, the male becomes female again (and the top of the hierarchy), the second-biggest female becomes a male, and they become the new breeding pair for that group.

    Biology is so weird/cool.

  22. #22 Laser Potato
    May 24, 2008

    That shrimp is no match for a killer KIKO-BEAM!!!

  23. #23 jck
    May 24, 2008

    Mantis shrimp are sort of like Republicans- they see a different reality than I do!

  24. #24 Laser Potato
    May 24, 2008

    For those who didn’t get that reference…
    http://www.albinoblacksheep.com/flash/kikkoman-e
    SHOW ME!!! SHOW YOU!!!

  25. #25 Steve
    May 24, 2008

    So if their vision is so great, what’s their spatial contrast sensitivity function for polarized light like? The electron micrograph shown in figure C has a scale bar but there’s no dimension attached by which to measure lens spacing.

    Could they discriminate the fine print of an insurance contract in low light if their life depended on it?

  26. #26 Wim L
    May 24, 2008

    You think God doesn’t hate shrimp? You just need to look at it from an appropriately Gnostic or Lovecraftian viewpoint. For God so hated the shrimp that he gave them eyes to see the true evil and horror of the world they live in. You should thank your Maker for your relative blindness!

    Seriously though shrimp vision is amazing: in addition to polarization, they appear to have possibly ten different color receptors. Not for them the lowly three-dimensional color-cube. Can you imagine trying to do art or graphic design in ten-dimensional color spaces?

  27. #27 mikespeir
    May 24, 2008

    I’d look even funnier with eyes like that.

  28. #28 andyo@yahoo.com
    May 24, 2008

    I think the mantis shrimp makes a brief appearance on Planet Earth, if I’m not mistaken. Anyway, a few days ago a friend of mine was asking me what was my favorite animal (I told her about the other half of the Homo sapiens, and the better-looking part of it at that).

    Anyway, she then told me she was asking what I would like to be reincarnated to (I know, I know) I told her after a bonobo, maybe the mantis shrimp cause it’s a damn cool animal. Not only the eyes, but using them for spear-hunting.

  29. #29 andyo
    May 24, 2008

    Oops, that above in my name is not my real email, just in case spammers are looking. I don’t know whose email is it. Spam away if you must.

    By the way, there are various videos of mantis shrimp on youtube. They seem to be pretty popular with the aquarium people too.

  30. #30 Laser Potato
    May 24, 2008

    I think Bronze Dog once asked an ID proponent basically “if we were designed so perfectly, why is our eyesight so poor compared to that of squid?”, then going on to list the flaws of the human eye. The guy’s response was that our eyes ARE perfect, we skeptics are just too dumb to see it. BD then went “but if they were perfect, they wouldn’t have ANY design flaws.” Predictably, the IDiot avoided the issue after this.

  31. #31 Epinephrine
    May 24, 2008

    We had a stomatopod in our reef tank several years back, amazing little things. Took out a lot of our snails before we knew he was in there.

    Read a little paper on their use of cavitation effects to amplify their blows. How cool is that?

  32. #32 Ichthyic
    May 24, 2008

    If anybody else is really interested in Stomatopods, Roy Caldwell up at Berkeley has been working on them for years (he was on my advisory committee when I was there).

    He’s pretty accessible, and loves to tell stories, too.

    http://ib.berkeley.edu/research/interests/research_profile.php?person=38

  33. #33 Jens
    May 24, 2008

    This is interesting from an evolutionary perpective. To think that despite not having the lineage to develop polarized light sensitivity we still have a nacent ability to sense it (Haidingers brush) that varies across the population. So, if humans were ever presented with the selective challenge of having to detect translucent or shiny prey we too might develop the ability to a greater extent.

    It sounds a bit like a subplot for a SciFi novel. I can see it now, “Humans have been subjugated for generations by a master race of nearly invisible alien creatures from a distant planet. In the year 3021 a hero is born able to see these creatures and sets out to find others with his unique powers…”

    I call first dibs on the movie rights! Somebody else can write the book.

  34. #34 sleepyinsaudi
    May 25, 2008

    Don’t cha know that if you wear your poloroid sunglasses, you can see where the lizard,I mean shrimp overlords are? I don’t leave home without them.

  35. #35 Wayne Robinson
    May 25, 2008

    I have just finished reading “The Eye A Natural History” by Simon Ings (and a very good read it is too), and he mentions the mantis shrimp too as having polarised light sensitivity and also has 16 different colour photoreceptors. He also mentions that in humans, pigment in nerve fibres in front of the macula (the portion of the retina most sensitive to fine detail) acts as a polarising filter, but whether we appreciate it I doubt. Simon Ings concluded his book by noting: “That is it not ironic, that in 538 million years of natural selection, eyesight should evolve from a simple light-detecting cell, pass through numerous variations and generate countless different ways of seeing, and come at last to serve as the dominant sense of the planet’s dominant species-an animal who only sees what it wants to see?

  36. #36 Etha Williams
    May 25, 2008

    @#35 Wayne Robinson –

    Simon Ings concluded his book by noting: “That is it not ironic, that in 538 million years of natural selection, eyesight should evolve from a simple light-detecting cell, pass through numerous variations and generate countless different ways of seeing, and come at last to serve as the dominant sense of the planet’s dominant species-an animal who only sees what it wants to see?

    I love it! I want to use that next time a cdesign proponentist starts in on the complexity of the eye….

  37. #37 Luke
    May 25, 2008

    So How many rods and cones do they have relative to humans?

  38. #38 Brandon P.
    May 25, 2008

    That’s isn’t the only way we Homo sapiens are exceeded by other species. There is an extinct African human species with bigger brains than us, as well as an African fish with a proportionately bigger brain than us.

  39. #40 JohnnieCanuck, FCD
    May 25, 2008

    There’s more at xavi’s link in #3 above than just human P-vision.

    The octopus section describes P-vision in octopuses and cuttlefish (yay cuttlefish!). Not only that, but cuttlefish, at least, can control the polarization of light reflected from their skin with their iridophores.

    All fascinating.

    http://www.polarization.com/

  40. #41 Spell Checker
    May 25, 2008

    Do you mean ommatidia, not ommatida?

  41. #42 Andrew
    May 25, 2008

    They also taste good, too!

  42. #43 agimat
    May 25, 2008

    Just for that, we’ll chow down on shrimp tonight!

  43. #44 Josh
    May 25, 2008

    I don’t think it would be a survival benefit to see in that sort of light, but it obviously is a reproductive benefit for the shrimp – the females need to be able to see which males are reflecting the best polarised light to see which is the fittest, and the males need to know who their competition is.

    When the scientists find a creature that can actually see, with their eyes, in the deep infra-red (~1000-5000nm at least), then I’ll turn green with envy.

  44. #45 John C. Randolph
    May 25, 2008

    Shrimp are my third-favorite arthropods, after dungeness crabs and lobsters. Yum.

    -jcr

  45. #46 David Marjanovi?, OM
    May 25, 2008

    When the scientists find a creature that can actually see, with their eyes, in the deep infra-red (~1000-5000nm at least), then I’ll turn green with envy.

    What about that extra organ that the viperid snakes have?

    There is an extinct African human species with bigger brains than us

    30,000 to 10,000 years old? In other words, last week. Can hardly be a separate species under almost any species concept.

  46. #47 Nick Gotts
    May 25, 2008

    Re #39 – the “extinct African human species with brains bigger than us” is “Boskop Man”, which scientific consensus views as artefactual. The book pointed to is discussed (not reviewed) here: (http://johnhawks.net/weblog/reviews/brain/paleo/lynch-granger-big-brain-boskops-2008.html).

  47. #48 silence
    May 25, 2008

    You know, humans actually can detect polarization with the unaided eye. We’re not very good at it, but it is possible.

    Look up Haidinger’s brush, and practice a little.

  48. #49 markg
    May 25, 2008

    That’s fascinating about the eyes. I’m afraid I’m more interested in the flesh, especially when coated in coconut and deep fried.

  49. #50 PeteK
    May 25, 2008

    Customer: Do you serve shrimps here?
    Chef: We serve anyone sir, sit down…

  50. #51 Spanky
    May 25, 2008

    Nobody could possibly “see” the phase, even classically speaking it oscillates at about a 10^15Hz rate for visible light and that’s a bit too rapid, even for shrimps. But classical picture is inappropriate here. The only thing you can observe is the relative phase of two beams, through the interference pattern, and it’s just the amplitude, anybody can see that.

  51. #52 Embolomere
    May 25, 2008

    OK, but what is the functional advantage to the critter in its daily life? What important things can it perceive?

    Put another way, what disadvantage would a beastie have without these abilities? What important things would it not be able to detect?

  52. #53 Darby
    May 25, 2008

    A few other things I’ve heard about mantis shrimp and are probably true-

    Their back-bent legs reach out at great speed to snatch prey with great force.

    If they hit the glass of an aquarium tank, it will usually break.

    Fisherman often call them “thumb poppers” because if you’re pulling them from a net, they can grab and break digits.

    They may not be shrimp, depending upon how you define shrimp. But then, most things called shrimp wouldn’t be.

    They have trinocular vision (concentrations of oriented fine-resolution ommatidia). In each eye.

  53. #54 Lynnai
    May 25, 2008

    I think being able to see polarized light (more then just hardingers brush) woul,d make looking at crystals even more interesting.

  54. #55 oscarzoalaster
    May 26, 2008

    It would be really cool to be able to add being able to see polarized light to the design of the human eye…. (along with ultraviolet vision, additional color receptors, and so on.) It would be nice to be able to see a rainbow in _all_ of its glory.

  55. #56 Josh
    May 26, 2008

    @Embolomere (#52), the researchers found that the male shrimp reflect polarised light. So likely a shrimp that couldn’t detect such light would be unable to read sexual cues, or at least wouldn’t know if another shrimp was hot or not.

    Oh, @David Marjanovi? (#46), the thermoreceptive pit organ is still a proto-eye, resembling a pinhole camera. No doubt any image is very poorly resolved. In that sense it is only slightly better than the human ability to feel the heat of a fireplace by sitting by it. I was referring to any animal which has proper photoreception for such long wavelengths of light, able to see IR light just as easily as green or red light.

  56. #57 Embolomere
    May 26, 2008

    Josh@56
    Thanks. Seems plausible.
    Oh, and I just reread the last quoted paragraph in PZ’s article:

    Stomatopods are shallow-water crustaceans in a visual environment with a partially polarised background. … This goes beyond simple contrast enhancement: optimal polarisation vision is analogous to the improvement afforded by stereo over mono vision in terms of increased information capacity.

    So it seems that in the ever shimmering and shifting light conditions of the shallow water environment, it would be like having super-duper polarizing shades, and thus reducing [eliminating?] the distracting and/or confusing effects of continuously flickering light. Useful for better prey and predator detection.
    Sorry, I should read more carefully before taking up more valuable blog real estate.

  57. #58 David Marjanovi?, OM
    May 26, 2008

    I’ve read owls can see into the IR, but probably that’s the very near IR.

  58. #59 TheBlackCat
    May 26, 2008

    I think being able to see polarized light (more then just hardingers brush) woul,d make looking at crystals even more interesting.

    I was reading a book called “Other Senses, Other Worlds” about how having different senses might impact an extraterrestrial species. The book wasn’t that good (“Extraterrestrials: A Field Guide for Earthlings” was much better). However, one example was a species that was sensitive to polarized light. They made crystal carvings that looked like nothing to humans normally but if looked at through a polarized filter showed different images depending on which way you held the filter.

  59. #60 Owlmirror
    May 26, 2008

    the thermoreceptive pit organ is still a proto-eye, resembling a pinhole camera. No doubt any image is very poorly resolved. In that sense it is only slightly better than the human ability to feel the heat of a fireplace by sitting by it.

    Actually, I recently read something about how the snake brain will do some high-level signal processing on the thermal image to improve the effective resolution. Let me see if I can dig that up…

    OK, this is what I was remembering:

    http://www.physorg.com/news76249412.html

    For whatever that’s worth.

  60. #61 Colin
    May 26, 2008

    I’ve been eating them for years in an attempt to gain their powers!

  61. #62 Greg
    May 26, 2008

    Correct me if I’m wrong, but doesn’t the photoelectric effect describe light as traveling in discrete photons, with energy directly proportional to the frequency. That is, the amplitude of light waves doesn’t vary, just the number of incident photons.

  62. #63 Josh
    May 27, 2008

    Greg, I think that’s an issue of the wave-particle duality of light. The amplitude of the light wave and the intensity of the light (number of photons per unit area) are interrelated. I don’t know how exactly…it’s all very confusing for a biologist like me.

  63. #64 Big Al
    May 27, 2008

    There is some evidence that the human eye does have some sensitivity to the polarisation state of light. Do a search for Haidinger Brushes.

  64. #65 whatisthis
    May 27, 2008

    science. I love it.

  65. #66 Mark
    May 27, 2008

    Well this kinda puts paid to that Intelligent Design canard about the human eye being much to complex to merely have evolved. By this reasoning we’re not god’s chosen ones, the shrimps are (for why else would he give THEM the better eyes!!)

  66. #67 PS3 guy
    May 27, 2008

    Shrimp can see things we can’t because they need to. God had the foresight to give us what we need and shrimp what they need.

  67. #68 Laser Potato
    May 29, 2008

    *glances at poster’s website*
    Great, a scam artist AND a creationist. Whoop-dee-freakin’ doo.

  68. #69 Jocelyn
    May 31, 2008

    This is eerily reminiscent of a paper that came out in March in Current Biology:

    “Circular Polarization Vision in a Stomatopod Crustacean”
    http://www.current-biology.com/content/article/abstract?uid=PIIS0960982208002522

    Roy Caldwell, mentioned by another commenter, is a co-author. Check it out!

  69. #70 erichansa
    July 25, 2008

    these eyes are likely as flydragon.