How can a colorblind animal change its colors to blend into the background?

May 16, 2011
This image shows a calibrated well-camouflaged cuttlefish. Credit: Photo by JK Wickiser, West Point.

How could a colorblind animal know how to change its skin color to blend into its surroundings? And what will the animal's predator "see," looking at its prey before and after it hides?

These provocative questions are addressed in article published today by a collaborative team from the Marine Biological Laboratory (MBL) in Woods Hole, Mass., and the U.S. Military Academy in West Point, N.Y. The article, "Hyperspectral imaging of cuttlefish camouflage indicates good match in the eyes of fish predators," appears in (PNAS).

"What makes camouflage effective? The answer is that it's an opinion, and that opinion completely depends on who is being asked. Our work presented in this paper takes the field a step closer to quantifying camouflage effectiveness," says J. Kenneth Wickiser, Asst. Prof. in the Department of Chemistry and Life Science at the U.S. Military Academy in West Point.

Understanding the principles of camouflage is not only important in biology, but also provides insights into architecture, advertisement, and defense applications.

For this study, the team studied camouflage in the known as coleoid cephalopods (octopus, squid, and cuttlefish). Camouflage is the primary defense in these animals and their rapidly adaptable body patterning system is among the most sophisticated in the animal kingdom. The expression of camouflage body patterns in cuttlefish is a visually driven behavior. Previous studies have shown that certain background variables—such as brightness, contrast, edge and size of objects, etc.—are essential for eliciting camouflaged body patterns. However, cephalopod eyes lack color perception, thus the vexing question of how they achieve effective camouflage while being colorblind still remains.

Camouflaged cuttlefish. ROI greyscale 2 white 2 black. Credit: J.K. Wickiser, West Point

Moreover, camouflage studies suffer from the inability to assess the effectiveness of camouflage in the visual space of predators; in short, it is difficult to determine whether a predator actually visually picks out the prey from the background environment. Knowledge of the visual capabilities of many predators is surprisingly lacking. Thus, the team asked, how does a colorblind animal change its color to blend into its surroundings, how do predators actually "see" their prey?

"To tackle this incredibly difficult problem, we put together a team with Chuan-Chin Chiao as a vision expert, Roger Hanlon and his scientists at the MBL as world leaders in cuttlefish and camouflage biology, and combined it with our expertise in hyperspectral imagery analysis at West Point," says Wickiser.

The team exploited a new imaging technology (HyperSpectral Imaging, HSI) to more accurately measure color match between animal and background, and to enable them to model camouflage in the eyes of predators. HSI employs a camera that captures not just 3 narrow windows (Red, Green, Blue – RGB) like human eyes or typical digital cameras do rather it captures the entire spectrum using 540 windows. In essence, HSI offers the opportunity to "see" things that humans, and many predators, cannot because of the limitations on our eyes. Because of the wealth of information HSI images provide, the scientists could extract a small amount of that data and compile an image from the perspective of a predator.

Importantly, modeling color vision of potential di- and tri-chromatic (RG or RGB capable) fish predators of cuttlefish corroborated the spectral match analysis and revealed that much of the contrast information (which allows a predator to "pick out" a cuttlefish from the background environment) resides in the brightness (luminance) rather than in the color (chromatic) aspect of the reflected light. What this means is that cuttlefish camouflage strategies take away a tool from predators in their ability to pick out their prey from the background and instead leave them with only brightness as a method for prey identification.

These findings (i) indicate the strong potential of HSI technology to evaluate camouflage body patterns simultaneously in the spatial and spectral domains, (ii) provide supporting evidence that cuttlefish can produce color-coordinated camouflage on natural substrates despite lacking color vision, and that (iii) the color aspect of cuttlefish camouflage is highly effective against fish predators. This is the first time that color matching in camouflaged animals has been visualized more realistically through the eyes of their potential predators.

"So much is unknown about how predators actually see their prey. Using hyperspectral imagery tools is a huge advance in getting us the information we need to model predator vision," says Wickiser. "We hope our work takes us one step closer to understanding how a colorblind animal adopts near-perfect camouflage in a variety of backgrounds."

The color-changing abilities of cephalopods have been appreciated since Aristotle's time. Although this topic has received much attention in the past, no quantitative assessment has been made to examine the color match between animal and background. Applying hyperspectral imaging system and modeling the predator's visual system adds a new dimension to quantifying animal camouflage in the eyes of the beholders. The team's findings provide strong support that cuttlefish are capable of hiding in plain sight of their visual . Furthermore, this approach may prove to be tranformational in the way that color is quantified in sensory studies of and signaling in the natural world.

Explore further: From sharks to baboons - insights into the wildlife wars

More information: Chiao, C-C, Wickiser, J.K., Allen, J.J., Genter, B., and Hanlon, R.T. Hyperspectral imaging of cuttlefish camouflage indicates good color match in the eyes of fish predators. PNAS Early Edition, week of May 16, 2011.

Provided by Marine Biological Laboratory

3 /5 (2 votes)

Related Stories

Stealth camouflage at night

Mar 09, 2007

Cuttlefish are well-known masters of disguise who use highly developed camouflage tactics to blend in almost instantaneously with their surroundings. These relatives of octopuses and squid are part of a class of animals called ...

Recommended for you

New hope for beloved family pets

11 hours ago

Nearly one out of four dogs will develop cancer in their lifetime and 20 per cent of those will be lymphoma cases.

User comments : 33

Adjust slider to filter visible comments by rank

Display comments: newest first

ibuyufo
4 / 5 (3) May 16, 2011
Maybe I'm blind but where exactly is this cuttlefish?
yOnsa
5 / 5 (4) May 16, 2011
lower right pie section of the circle, i think. see those bumps that aren't as random as the rocks?
ibuyufo
5 / 5 (2) May 16, 2011
Thanks! Gotta hand it to the cuttlefish!
Eikka
4.3 / 5 (4) May 16, 2011
Considering that the fish live in the same environment with the same type of rocks and sand for eons, I don't think it's far fetched that they simply take a guess based on what surface they happen to land on.

Life quickly weeds out those who guess wrong.
hush1
1 / 5 (1) May 16, 2011
...the way that color is quantified in sensory (perceptions)studies of camouflage and signaling in the natural world.


Close enough. Have fun rediscovering centuries-ago-answered questions. Just don't get caught rehashing the past as something new. You might lose cover.
DavidMcC
not rated yet May 17, 2011
This is a very interesting conundrum. I wonder whether cuttlefish (and octopuses/octopodes) express visual opsins in patches of their skin that may (directly or indirectly) connect to their chromatophores. I can't think of any other way they could do this camourflage trick. Such additional opsins would not be expressed in their eyes, but could still be identified in their genome. I do not know of any research results on this.
PS, I thought it was in the bottom left, not the right. Could be wrong though. :)
DavidMcC
not rated yet May 17, 2011
... It would make sense to take the colour vision out of the imaging eye if all that is required is the general colour of the background. The reason is that there would otherwise be a significant price to pay in terms of acuity, due to the limited space available in the retina for the extra colour receptors.
DavidMcC
not rated yet May 17, 2011
Hush1, I take it that you have references from several centuries ago on the biology of cephalopod camourflage!
antialias
not rated yet May 17, 2011
The interesting thought in this article is that it's not the camouflaged animals' vision that matters but the predators' vision which must be fooled. However, the prey must still know which types of background will be efficient camouflage.

It seems to me that this would force a convergent evolution of the eyes of predator and prey (at least of the color spectrum they are sensitive to). Any maladatptaion would lead to a wrong choice of camouflage and therefore to sudden death.
DavidMcC
not rated yet May 17, 2011
Not necessarily, antialias. The best strategy when there are diverse predators (probably with varying vision) is to simply match the skin colour pattern to the background over as wide a range of wavelengths as possible.
____
On the question of control of chromatophores, there are at least three modes that cuttlefish have: camourflage, courtship and other signalling. I suspect that this means that the only conscious control the animal has over the chromatophores is which mode to be in.

PS. I've just seen it in the picture! Right across the bottom.
DavidMcC
not rated yet May 17, 2011
For the camourflage to work, there would also have to be input fom the creature's eyes, just to get the texture correct (minus colour), combined with colour sensing from other skin organs than the eyes.
DavidMcC
not rated yet May 17, 2011
It seems that the opsins expressed in the skin of cuttlefish have already been found:
http://rsbl.royal...223.full
"Evidence for distributed light sensing in the skin of cuttlefish, Sepia officinalis"
So, conundrum solved (albeit only last year)!
hush1
not rated yet May 17, 2011
The article is somewhat misleading. How do you represent a colorless world when your only perspective (senses or perceptions) are that of a cuttlefish? The spectral discrimination has been eliminated. So a background with color needs no further consideration. The interactions or exchange of information between cuttlefish and environment does not include spectral discrimination. What are the signal representations of recognition and operation (of living) in a colorless world? The organism is able to perform without spectral discrimination. Yet the authors opt for this.

I have century-old references to signal representation in a colorless world. Do those signal representations (energy distributions) have a biological/chemical expression in cuttlefish? Assigning 'purpose' to expression can be done at anytime during research.

People born blind and color blind have representations in their mind of what color is. A different representation from ours. What is 'camouflage' to them?
DavidMcC
not rated yet May 18, 2011
Hush1, check the link I supplied above. It shows that cuttlefish do actually sense the colour of their environment, but not using their eyes. This is because space on the retina is too valuable to waste on something that can be done in the skin. It doesn't give them colour vision, but it does enable them to use colur to camourflage themselves effectively.
I doubt that this had been fully understood a century ago.
DavidMcC
not rated yet May 18, 2011
Sorry, repeat comment. Due to a technical problem with the brouser or site.
DavidMcC
not rated yet May 18, 2011
Hush1, on the subject of "signal representations", I doubt that the ability to imagine colour would help a cuttlefish actually match its colour to its surroundings.
hush1
not rated yet May 18, 2011
Hush1, on the subject of "signal representations", I doubt that the ability to imagine colour would help a cuttlefish actually match its colour to its surroundings.


I agree. I share your doubt.
Obviously, the signals have a "representation" and form that differs from the point of origin where the signal originates.
(Maybe the 'depot' for this signal/information is the brain).
Whatever that brain "representation" is, the "spectral" discrimination is not "spectral" for the cuttle fish.

To end speculation, simple put cuttle fish in an artificial "black and white" environment and watch what happens.
The smallest and cheapest test to big research.
DavidMcC
not rated yet May 19, 2011
Hush1, speculation on this subject has already been ended by the work of Mathger et al. (linked above) that showed that cuttlefish can detect the colour of their surroundings through visual opsins in their skin, as I mentioned before.
hush1
not rated yet May 19, 2011
The work of Mathger et al. ends all speculation on this subject!

(!) <--(eyebrows raise, eyes widen, jaw drops in dismay)

DavidMcC
not rated yet May 20, 2011
That's right, hush1. There is absolutely no need to propose that random guessing is required when the cuttlefish can detect colour in its surroundings, even if it isn't actually seeing the colour. The chromatophores that determine the colour pattern in its skin can be adjusted using the non-eyesight colour information.
DavidMcC
not rated yet May 20, 2011
... It seems that there was no room in its retina for all the extra photoreceptors, which would degrade its visual acuity unless its eyes got a lot bigger (which would be expensive). Perhaps it doesn't use them for hunting, only for predator avoidance by an unconscious camourflaging process.
hush1
not rated yet May 20, 2011
Put the fish in a black and white environment. Regardless of God's absolute...oops wrong thread,... regardless of the work of Mathger et al. that ends all speculation. Nothing is absolute.

And this fast, cheap, simple, experiment can be just another feather in the cap/hat of additional data fortifying the researchers conclusions. 100% confidence researchers welcome anything strengthening an original hypothesis/theory - the more additional data confirms, the more merrier the researchers. Nothing can "threaten" a confidence level of 100 percent. So why even the need for additional experimentation to support a successful existing theory?

That was stated already. Nothing is absolute. There are as many rationalizations for not questioning the "end of all speculation" as there are the number of questions, questioning the exist of God.

When I last checked, that was a lot questioning.
And we all know God needs no data or support.
DavidMcC
not rated yet May 23, 2011
Hush, stop twisting my words about "100% confidence" (your words, not mine) to cover for a falsfified hypothesis. There is no way that the superb camourflage colouring that cuttlefish (and some species of octopus) consistently achieve within seconds is done without using the colour receptors in their skin.
Basically, it's absurd to portray a scientific conclusion as being religion-based, as you are doing, on the basis of a phrase I used.
DavidMcC
not rated yet May 23, 2011
To put it another way, hush, it is bad science to cling to a falsified hypothesis that relied on lack of information at the time.
DavidMcC
not rated yet May 23, 2011
BTW, hush, I'm an atheist!
hush1
not rated yet May 23, 2011
There are as many rationalizations for not questioning the "end of all speculation" as there are excuses for not questioning work at all.

Here's one more:
"There is NO WAY ... in their skin."

You missed the point referencing God completely. The emphasis is on the RATIONALIZATIONS anyone uses to defend conclusions.

Whether those conclusions are reached through scientific method or religion is completely irrelevant to the point at hand - namely, your rationalizations.

"...NO WAY..." is a rationalization. A completely unscientific, unfounded, illogical expression to avoid scientific inquiry and legitimate question or doubt.

Alright. I'll state this differently:
Nothing is absolute. There are as many rationalizations for a scientist's 'fact' as there are falsehoods. Every one of those rationalizations make "sense" for a scientist as long as those rationalization support the scientist's fact and not the falsehood.
DavidMcC
not rated yet May 23, 2011
Hush, once again, you are turning truth on its head. There is, indeed, no way that any organism can respond reliably to its environment without information about that environment.

You, presumably, think that the late Paul the octopus actually had a supernatural ability to predict the result of football matches!
DavidMcC
not rated yet May 23, 2011
... Of course, I am not arguing that Paul the octopus was using his senses to predict football match results, only that, because he could not know anything about football matches, that he could not have been involved in any overcoming of heavy odds against success that might have occurred.
hush1
not rated yet May 23, 2011
If you want to label scientific conclusions to experimentation "truth", by all means, do so.

"There is, indeed, no way that any organism can respond reliably to its environment without information about that environment."

We, humans, respond reliably to death. There is no information we know of, about that state of the environment.
DavidMcC
not rated yet May 24, 2011
Hush1, I'm afraid "death" isn't an object in the environment, so your argument is nonsense - we do not "respond to death", we simply die. Although this ridiculous thread seems to go on forever, largely because of your obsession with a bit of century-old woo science.
By "truth", I mean an observable fact.
DavidMcC
not rated yet May 24, 2011
...And yet another point, hush1. Your use of the word "rationalization" is what actually applies to your pet woo theory, because the word implies bending logic to make what does not make sense SEEM to make sense. The Mathger paper allows us to make sense of observations, your woo forces rationalizations (ie nonsense, such as "they imagine the colours").
hush1
not rated yet May 24, 2011
You are sadly mistaken. The outcome of dieing is death. A part of Nature. If you do not view this as a part of Nature, you are not a scientist. Which you have supply amply wording to support. If you are anything, being a scientist or conducting science is not what you do.

"death" isn't an object in the environment


Enlighten us with your death-defying semantics of what death is.
You have wasted everyone's time.

Anyone can see this "fact".
It is "observable" right here.
I see no point to further engage in your nonsense.
Feel free to "woo" others with your "logic". (Wave your wand)

Let us know if your work even does ever turn towards science.
DavidMcC
not rated yet May 25, 2011
This thread is getting silly. Hush1, the most naive person on the site, who understands nothing about science, is pretending he's the expert and I'm just a fool.

Perhaps physorg staff should indicate a final score, by actually rating the posts?