Modeling Bee Vision with Stacks or Filters? Well, no, we can't really.

[Andrea B.]

Preface: I would never, ever suggest you stop making enjoyable and interesting photographs which represent the wavelengths which bees can detect. But please do learn and understand that current bee vision researchers will never accept our photographs or photo stacks as anything resembling a good model of bee vision. Because they are not even close.

 

This does not mean at all that we shouldn't continue to discover and appreciate what we find in our alternate wavelength photogaphs. UV-signatures, for example, are there for a reason even if we don't know for sure what that reason is.

 

As Professor Horridge says: "No apology is needed when new facts interfere with old beliefs".

Further, "100 years of conclusions need serious revision because they were based on incorrect theory."

 

**********

 

 

The few notes and references I'm going to list below are only a very tiny portion of the whole complex thing that is Bee Vision. Also note that many research papers in the last few decades on Bee Vision are now sadly out-of-date in light of recent studies. And many of the discussions which I and others have made here on UVP are just plain wrong. That's OK, I don't mind, that's how we learn.

 

((I'm not sure I can work through this entire website and edit incorrect references though. So let's just move forward.))

 

NOTES:

 

Bee vision is quite counter-intuitive compared to human vision.

 

Bees do not see colors as color, per se. Think of it more like this - the bee detects wavelengths in UV, green, blue.

 

Bee vision is not trichromatic.

 

In each bee ommatidium (area of the compound eye) there are

• 6 green sensitive cells
  
• 1 UV sensitive cell
  
• 1 blue sensitive cell
  

.

UV: The UV receptor is used by a bee to detect the direction of the sky for level flying or for escape. UV plays no role in bee foraging. UV is not really a colour to the bee as we humans think of color.

 

Blue: Bees locate and measure amounts of (tonic, steady-signal) blue in areas.

Bees detect differences in blue content relative to the average blue for the background.

 

Green: Bees separately measure quantities of (phasic, while in motion) green contrast at edges.

Location of the green contrast cue in a particular area of the bee eye is important.

 

Bees also measure the angle between blue content and green modulation.

If blue content and green contrast cues are not present, bees can measure blue modulation.

Orientation detectors are restricted to the green channel.

 

White: To a bee, white is an intense blue.

 

Black: To a bee, black is zero blue with maximum edge contrast.

 

Bees distinguish between black, white and grey cues by measuring changes at edges and not by contrast alone.

 

 

Blue is a preferred input for bees.

Bees are not normally attracted by green or black

Colors are not located or separated.

Pattern layout is of not interest or not detected.

Total blue content and vertical edges were detected and learned retinotopically.

There is no evidence that green content is measured or related to blue content.

 

Bee vision is not trichromatic, but UV inhibits the effect of blue content in white.

 

 

Monochromatic detection advantages:

 

Monochromatic blue receptor in large angular field measures greater or less blue content relative to bkg reduces noise (because monochromatic) and discovers targets (flowers).

 

Monochromatic green receptor in small angular field detects landmarks, motion, optic flow and stabilizes the eye, controls the flight path, and measures range (fom size & angular velocity).

 

 

***********************

 

These notes are mostly from the article Bee Vision is totally different by Adrian Horridge (Australian National University) in the journal The Australasian Beekeeper, 2017, Vol. 118, No. 12, page 544.

 

Professor Horridge makes all his work, including the referenced article, freely available here: http://www.Adrian-Horridge.org

In particular see: How bees distinguish colors.pdf

 

Professor Horridge's work has been mentioned before in various posts here on UVP. I am not the first to make references to his papers.

 

---

---

 

ADDED 18 May 2018

Editor's Statement: The floral UV-signature is real. It exists. It is associated with the underlying pigment chemistry in the flower. That a bee does not interpret the UV, blue and green reflective signals from a flower face in the way that was originally postulated does not negate the existence of the pigments and the reflective properties of the flower.

 

All along I've tried to point out the the UV photos we record are only one channel-ed. For any creature having UV visual receptors, the UV photograph represents the possibilities -- but not necessarily the realities -- of their very complex vision.

 

 

---

---

[DaveO]

Here's another view of the subject:

 

More than colour attraction: behavioural functions of flower patterns Natalie Hempel de Ibarra, Keri V Langridge and Misha Vorobyev

 

Current Opinion in Insect Science 2015 12:64-70

 

Abstract

Flower patterns are thought to influence foraging decisions of insect pollinators. However, the resolution of insect compound eyes is poor. Insects perceive flower patterns only from short distances when they initiate landings or search for reward on the flower. From further away flower displays jointly form larger-sized patterns within the visual scene that will guide the insect’s flight. Chromatic and achromatic cues in such patterns may help insects to find, approach and learn rewarded locations in a flower patch, bringing them close enough to individual flowers. Flight trajectories and the spatial resolution of chromatic and achromatic vision in insects determine the effectiveness of floral displays, and both need to be considered in studies of plant–pollinator communication

 

Conclusions

Pollinating insects forage in a three-dimensional environment and look at flowers from different distances and directions. What they see depends on the spatial resolution of the compound eye and visual mechanisms that process object information, however, it is also influenced by their flight trajectories and viewing conditions. What they choose depends on their vision and visual learning capabilities and is strongly influenced by navigation and spatial learning mechanisms. It remains to be understood how decisions are made and behavioural responses coordinated at far and near distances, as a pollinator moves between flowers, approaches and visits them. The underlying neural mechanisms involve basic sensory and motor systems that are shared across different taxonomic groups of insects. A wide range of flower search and choice behaviours adopted by insects can be explained by mechanistic models that take into account constraints imposed by the optics of insect eyes and aerodynamics of insect flight, rather than by models based on the assumptions of higher order cognitive processing of visual information

 

Here's the important figure

 

Dave

[Andy Perrin]

It certainly seems like it's a mistake to assume the UV bullseye pattern plays any role in flower selection if they can't even see it till they are within a few cm of the flower! The alternative flower would have to be immediately adjacent.

[Adrian]

A really interesting thread- I will add my own thoughts, for what they are worth.

 

I have recently started to give talks to local camera clubs and natural history societies on "Colour in Nature" (including "invisible" colour) causing me to give this some thought.

 

A few thoughts - human colour vision is very difficult to discuss (at a basic level) There is the old philosophical question- is my blue the same as your blue? Apparently not - you may remember a couple of years ago an image of a blue and black dress which lots of people saw as white and gold.

 

Many humans are colour blind, or deficient. There is some evidence that people in different parts of the world "see" colour differently, and I believe Kodak used to produce colour slide film in different colour balance for different regions of the world.

 

If we apply the same thought to insects, then it is possible that insects of the same species, may also have different perceptions of colour. If we just look at bees, there are hundreds, thousands of different species, all of which presumably have different colour sensitivity.

 

I think it is still valid for us to produce "bee vision" images, with the proviso that they are merely an indication or clue, rather than definitive answer to the question.

[Cadmium]

So do we then throw out your poster showing how the bee might see?

http://www.ultravioletphotography.com/content/index.php/topic/2598-comments-needed-for-this-poster-update-20170309/page__view__findpost__p__19588

 

I know your poster was calculated mathematically, so to speak, rather than a filter stack... and I know the filter stacks are not the correct mix of UV with visual.

The UG5 stack is UV + Blue + Green, but the UV is RGB, so false colors of the UV are being mixed with the Blue and the Green.

The stack looks like your poster. Accidentally, but accidental for just one kind of flower?

How does the stack shot compare with the 'poster view' for most other flowers?

[Andrea B.]

Editor's Statement: The floral UV-signature is real. It exists. It is associated with the underlying pigment chemistry in the flower. That a bee does not interpret the UV, blue and green reflective signals from a flower face in the way that was originally postulated does not negate the existence of the pigments and the reflective properties of the flower.

 

All along I've tried to point out the the UV photos we record are only one channel-ed. For any creature having UV visual receptors, the UV photograph represents the possibilities -- but not necessarily the realities -- of their very complex vision.

---

 

Cadmium, yes I might throw that poster out. I'm thinking what to do.

 

How does the stack shot compare with the 'poster view' for most other flowers?

I've only looked at a few. For any flower you need to figure out the actual reflectivities and compare them to what is being recorded in the raw file before making a white balance. Yellow flowers are particularly troublesome. Why does the filter recording yellow wavelengths as green? The filter does not appear to extend into the tiny yellow band.

[Andrea B.]

Dave O's paper reference: The abstract and conclusions certainly are in line with the Horridge model of bee vision as I currently understand it.

 

"Insects perceive flower patterns only from short distances when they initiate landings or search for reward on the flower."

This statement coorelates to the multiple small angle visual field landmarks made using green modulation detectors on edges.

 

"From further away flower displays jointly form larger-sized patterns within the visual scene that will guide the insect’s flight."

This statement coorelates to the large angle visual field landmarks made by measuring blue content of flower and background.

 

"...constraints imposed by the optics of insect eyes and aerodynamics of insect flight, rather than by models based on the assumptions of higher order cognitive processing of visual information"

This is exactly what Horridge is trying to emphasize. The bee brain is simply not large enough for higher order cognitive processing of visual information. The bee brain cannot hold a 'recording' of an entire flower or flower field. The bee brain is only large enough to hold a few blue content landmarks, green modulation landmarks or other small landmarks.

 

I think the authors of the paper veer slightly off the Horridge model when showing the projections of the flowers onto the bee eye facets although they do reference the small & large angles of the visual field. I might have attempted to translate those projections into information about blue content (for the 16° interpretation) and green edge contrasts (for the 7° interpretation). And don't forget there are symmetry preferences amongst other bee vision details.

[Andrea B.]

Adrian: There is the old philosophical question- is my blue the same as your blue? Apparently not - you may remember a couple of years ago an image of a blue and black dress which lots of people saw as white and gold.

 

Adrian, please do know that my response here is not directed at you, OK? :D

 

I think that fundamentally those of us humans having trichromatic color vision (i.e., not colorblind) see colors the same IF viewed in the same context. We all have the same pigments behind our visual receptors. And we all have the same brain structures. We all must be taught a name for any particular reflected wavelength we are looking at yet we all manage to assign the same color names to the same reflected wavelengths when viewed in the same situations. So I think it is biologically sloppy of many people to veer off into discussions of whether 'my blue is like your blue'.

[[That paragraph is my opinion based on what I've learned. Might change in the future, who knows? The word "same" should not be over-interpreted, please.]]

 

HOWEVER ---- we cannot ignore the side effects of our trichromatic vision processing. There are very real color perception phenomena (optical illusions) based on context which cause us to interpret a color reflection differently than its measured reality.

 

Somewhere there's a nice web page for colour illusions. I'm going to try to find it and post the link here: LINKIE_TBA

 

---

If we apply the same thought to insects, then it is possible that insects of the same species, may also have different perceptions of colour.

 

No, within those bee species which have been studied, anatomy and various experiments have shown they all process the same way. (Of course, some individual member of a species might have a mutation. But that poor guy is probably dead because it could not find food.) The experimenters have even wired up the little bee brains and artificially stimulated them to get similar responses to the same stimulations. (I don't like to think about this though because I love bees and anthropomorphically extend to them qualities which they most likely do not have amongst which are a horror of having one's brain wired up.)

[Jim Lloyd]

Many thanks for this, Andrea. I have read those papers, but will need to do so again a few times I think to properly understand.

 

I have been thinking for a while that in UV photography we are missing out on motion, and in trying to understand the way the bee finds the flower we need to think of its motion. So I have been looking at this for a little bit. It seems that a lot of the times bees and flies and going about fairly haphazardly. I saw a fly around a leopard's bane flower this morning - it seemed not to see it until it was a few cm away and then it looked like a magnetic attraction drawing it rapidly to land on one of the petals (rays). It then orientated itself and headed, head first into the flower centre to feed.

 

I have tried to film this sort of behaviour. See here . Its only a little clip and I guess far better has been done by professionals, but I find the approach of the hoverfly at the start fascinating to watch (appears after a few seconds).

 

Not sure what this adds ...

[Andy Perrin]

One thought is that the bullseye patterns, while irrelevant to the insects until they get very close, might have significance for other UV-seeing pollinators (hummingbirds?? etc.). Or perhaps they do serve some purpose for the insects at close range.

[Andrea B.]

Vision and Art: The Biology of Seeing

Revised & Expanded

by Margaret Livingstone

 

I've mentioned this book before. It is an enjoyable high-level discussion of human vision by a visual neurophysiologist. Touches on physics, human biology, optical illusions, art and more. Good diagrams and illustrations.

 

Livingstone devotes a full page (pg 46) to Do you see red like I see red? Her answer is basically 'yes' at the physiological level, but adds many nuances to that by discussing memories and life experiences which influence us. Livingstone concludes by regarding the original question as a semantic issue. What do you mean by it? Do you mean seeing red physiologically? Do you mean seeing red experientially? It's a good discussion.

[Andrea B.]

Here is an example of one-channel filtration for human vision. Suppose you had a nice daisy flower, white with a yellow center. If you photographed the daisy's red-signature, green-signature and blue-signature, here's what you would get.

Kindly grant me some leeway here. I used the photoshop filter tool on my fake daisy. With appropriately narrow bandpass filters, you could produce similar results in actual photos of a daisy.

 

If you were to pick only one of the following signatures, what could you conclude about human vision?

 

 

A Daisy in a Green Field

The yellow is (244, 244, 0). The green bkg is (85, 196, 0). I tried to make it a real grass color.

 

The Daisy's Red-Signature

 

The Daisy's Green-Signature

 

The Daisy's Blue-Signature

[Alex H]

UV: The UV receptor is used by a bee to detect the direction of the sky for level flying or for escape. UV plays no role in bee foraging. UV is not really a colour to the bee as we humans think of color.

 

My immediate question is - in UV plays no role in bee foraging, then why did so many unrelated clades of insect-pollinated plants evolved to have floral patterns visible only in UV spectrum? What is the alternative reason for it to be? What its evolutionary significance than?

 

I will read papers by Horrige later on, but it is strange he does not refer to any research except for his own.

[Andrea B.]

Horridge refers to the original bee experiments by von Frisch and vos Hess: Bee Vision is totally different

 

He refers to the work of the following other insect vision researchers in his book What Does the Bee See:

 

 

I owe a particular debt of gratitude to the numerous enthusiastic colleagues who made good use of the plentiful equipment, made scientific advances and mostly published their own work on insect vision. Their names appear throughout this book.

Some, in particular, stand out: Tudor Barnard, Malcomb Burrows, John Scholes, Steve Shaw, Rick Butler, Ben Walcott, Ayis Ioannides, Ian Meinertzhagen and David Sandeman.

In Canberra: Simon Laughlin, Allan Snyder, Andreas Dubs, Randolf Menzel, Gert Stange, Benno Meyer-Rochow, Ted Maddess, Fred Doujak, Peter Lillywhite, Jenny Kien, David Williams, Roger Hardy, Yasuo Tsukahara, Keiichi Mimura, Mark Leggett, Dan-Eric Nilsson, Mike Land, Stjepan Marčelja, Willi Ribi, Martin Wilson, Richard Payne, Roger Dubois, Doukele Stavenga, David O’Carroll, Tom Matic, Shi Jian, Jan Dalczynski, Peter McIntyre, Qijian Sun, Yang En-Cheng, Danny Osorio, Joe Howard, Mandyam Srinivasan, Tony Heyes, Andrew James, Zhang Shaowu, Miriam Lehrer, Eric Warrant, Mike Ibbotson and Neville Fletcher.

 

And there is 41 page Bibliography in the book of which only 3-4 of the pages refer to Horridge.

[Alex H]

I was referring to this paper: Bee Vision is totally different by Adrian Horridge (Australian National University) in the journal The Australasian Beekeeper, 2017, Vol. 118, No. 12, page 544.

 

The book is much broader in scope, of course it will have huge bibliography, but how much of it is relevant to this particular topic. Still, if bees do not use UV for foraging, why flowers evolved to have UV patterns like that?

[Andrea B.]

Well, Alex, I certainly do not know the answer to that. And Horridge does not address it. Horridge simply says that the UV receptor has not been studied in depth. What is know about the UV receptor so far is that it is critical to the bee for determining where the sky is so that it can orient itself in flight and also escape towards the sky as needed.

 

For what it is worth to any readers of UVP, I am entirely on the side of evolution myself. I do not think there is anything in nature that is not there for an evolutionary reason. Although we must grant that it sometimes takes a while for evolution to catch up if regarding the uselessness of the human appendix which still seems to hang around and cause trouble.

 

Everyone seems to be focused on the UV-signature and the bee, but as I just wrote somewhere else the significance of a UV-signature might also have benefit to the plant itself. Indeed I am recalling some paper somewhere (sorry can't remember just at the moment) which has discussed the underlying pigments of plants in regards to protection of the reproductive parts.

 

Anyone who does some further investigation about a UV-signature having benefits to the plant should search on flower pigments. There is a really fascination literature out there about that. (By 'literature' here I mean in the scientific sense.) I have referenced a nice book about plant pigments also.

[Andy Perrin]

Ooh, this is a good point (about benefit to the plant). Humans in sunnier regions have more melanin, so something analogous in plants seems plausible anyway.

[JMC]

So plants evolve sunscreen to protect their sexual organs. Those sunscreen absorb UV, so look dark to animals that can see UV. Plants also develop nectar to lure in insects to help with pollination. The nectar happens to be in the highly UV absorbing region of the plant. Insects associate food with the darker parts of the plant. Plant gets two benefits - UV protection and pollination.

 

I love being a simple chemist. Chemistry is a much more simple world than biology :)

[Alex H]

So plants evolve sunscreen to protect their sexual organs.

 

Now lets compare where the actual reproductive organs are, and how far the UV-absorbent areas extend to the petals/ligulas in something like Ranunculus or Bidens.

[Andrea B.]

Well before we all go too far over the cliff, let's find a good reference paper or two about possible benefits to the plant.

[DaveO]

From one simple (ancient) chemist to another (who has just posted the Bayer results that this simple chemist has long dreamed about). Your summary seems to me to be dead on target. Remember that ALL our antipodean flowers evolved without the benefit of European honey bees which have been used for all the classical bee-vision studies. So, the findings do seem to have a very basic fundamental truth, convergent evolution or whatever. There are sneaky Australian orchids that use pheromone attractants very similar to those of the pollinating wasps or other insects. We clumsy humans only think we know how the world works!

 

Here's a link to my musings about the UV signature of yellow acacias

http://www.ultravioletphotography.com/content/index.php/topic/1105-yellow-pigments-of-australian-acacias/

 

I suspect that similar compounds are responsible for UV absorbing signatures in other flowers

 

Dave

[Jim Lloyd]

My immediate question is - in UV plays no role in bee foraging, then why did so many unrelated clades of insect-pollinated plants evolved to have floral patterns visible only in UV spectrum? What is the alternative reason for it to be? What its evolutionary significance than?

 

I will read papers by Horrige later on, but it is strange he does not refer to any research except for his own.

 

This is a nice paper summarizing various findings on animals colour vision in relation to evolution. Maybe it's a bit dated (2001), but the authors conclude that for most animals UV detection is no more important than detection in other wavebands. They show that evolutionary analysis suggest that UV sensitivity is ancient (>350 million years old) and has undergone relatively little evolutionary fine tuning. They also show that pure UV reflecting flowers are virtually absent in nature. They make the general point that studies of animal colour vision should involve the whole of the range that the animal is sensitive to, and also consider the way the basic input signals are combined. For the bee they reference work that indicates that whether UV reflectance increases or decreases detectability depends on reflectance in other parts of the spectrum.

 

Quoting their conclusion:

 

 

 

"We conclude that there is little evidence to suggest that

ultraviolet is a special waveband for the visual systems of

animals and its importance has often been exaggerated (Kevan,

1972; Kevan, 1978; Kevan, 1979a; Kevan, 1983; Chittka et al.,

1994; Lunau, 1992; Kevan and Backhaus, 1998). Ultraviolet is

merely one component of visual stimuli that might be of

interest to animals (flowers, fruits, mates, enemies, escape

routes,oviposition, substrates,etc.).Although certain wavebands of light do sometimes have special significance to

some animals, ultraviolet wavelengths do not appear to be any

more significant than other wavelengths. These results suggest

that caution is required in interpreting the significance of

ultraviolet signals in the absence of information from other

regions of an animal’s visible spectrum. For the meaningful

evaluation of most animal behaviours, and especially where

colour vision is concerned, we argue that the salience of

ultraviolet signals is indeed limited. Investigators interested in

how animals visualise objects in their environments, and how

those objects might appear to animals, must exercise caution

and place ultraviolet radiation into its appropriate contexts."

[Alex H]

Jim, I fully support the statement that the UV reflectivity must never be considered on its own, but always in the combination with other factors. Unfortunately, what you write does not answer my question.

[Jim Lloyd]

Hi Alex

 

I am very much a beginner in this subject, so I don't know the answer to your question.

 

Maybe its not relevant, but the paper referenced by Chittka et al (1993) shows that only a small fraction of flower colours are in the bee UV, UV-Green or UV-Blue perceptual space. So I thought this lent weight to the argument that the UV signature was not very relevant in bee-flower evolution.

 

I am sure there are others on this forum that can give far better answers, but I like to try to learn by engaging in discussion.

[Andrea B.]

And I agree that it is a wonderful way to learn!

---

 

FWIW, I have these summaries for the Chittka group paper.

 

Notes: of the flowers surveyed:

• There are no UV-reflective black flowers.
  
• There are no cyan flowers.
  Note: Some blue flowers reflect a mix of +B+g, that is, more reflected blue than green.
  
• Green flowers are very rare. That is, red absorption is very rare in flowers.
  
• UV reflection is always lower than in other parts of the spectrum.
  

 

List: Percent of surveyed flowers = Bee Color reflected in some area of the flower.

Here "Bee Color" means only that the bee has a visual receptor for that wavelength area. I'm leaving out how the bee might actually detect a certain flower based on various models of Bee Vision.

 

Please observe that 96% of the flowers surveyed have some area which reflects blue, green or a combination of blue and green. This would seem to indicate that UV-reflectance, per se, is less important to the flower or to a flower + bee system? Just floating a hypothesis here.

• 04% +UV
  
• 11% +UV+Blue
  
• 12% +UV+Green
  
• 17% +Blue
  
• 17% +Green
  
• 33% +Blue+Green
  

 

List: UV-absorbing versus UV-reflecting

The larger percentage of UV-absorption seems to tell us that it might be more important to the flower or to the flower+bee system.

• 68% of the flowers surveyed had UV-absorbing areas.
  
• 25.6 of the flowers had UV-reflecting areas.
  

I might have posted this in the past here or on the old NG UVIR-board.