a good comparison is to how floating point arithmetic works. computers are actually terrible at math, because they have to estimate everything they do and carry along error with them at every stage of the calculation. a programmer is keenly aware of this. now, if a computer is unable to come up with the same answer to the same arithmetic problem twice, does that mean the arithmetic problem has multiple answers? or that the computer is perceiving the problem differently? no. it's just wrong.
there's an opportunity for humans to be humble here, in understanding how fragile and poorly evolved we really are. but, we need to drop the bullshit, first.
if you can't see the right colours under situations where you have a proper control in place, it's because your eyes suck. they're not able to process what exists in front of you. it's a limitation of your senses, not some kind of evolutionary advantage.
Tessa Townsend (quintessential)
+deathtokoalas What are the correct colours in your opinion? (I'm just interested) because the dark blue and black are the correct colors of the dress in reality but the image has neither dark blue or black in any pixel does it?? I do think this dress points out a flaw in our perception systems it's really interesting :)
deathtokoalas
+Tessa Townsend i had no problems seeing the dress as it appeared in the rgb analysis. the dress is not the picture of the dress. the dress is black and blue. the picture of the dress is purple/blue and brown/orange. there's obviously some grey area there in terms of what you call a colour. but it's not white, and the question that needs to be answered is why the people that see it as white are unable to pick up the colour in it.
"washed out blues" are common with conditions like cataracts and general age-related vision loss. and, what people are saying is generally consistent with the idea that they have some mild colour-blindness. it could be damage from too much tv, too much time in the sun, etc. it could just be genetic. i don't think you're looking at a single cause of this vision failure.
i'm just irritated by the way people are trying to explain this, and trying to pass it off as "science". almost nobody seems able to articulate the right question, so it's not surprising that almost nobody seems to be able to come up with the right answer.
Tessa Townsend (quintessential)
+deathtokoalas And I really like your computer analogy I had to read it twice to understand it (I'm terrible with computer theory), but my mums an IT teacher and your new biggest fan.
I was one of those who saw white and gold at first, and if you had asked me if it was really pure white I would have said no its light blue but you can tell its really white because white things often go light blue when you take photos of them REF : http://digital-photography-school.com/why-is-the-snow-in-my-pictures-so-blue/ I mean it can't be just a problem with vision an the lack of ability to see certain colours in clarity because about half of the people on this comment thread have said the dress swaps every time they look at it. You're right we're, (and by we're I mean the scientific community), trying to come up with an answer to a question in 24 hours, I mean scientists spend a whole lifetime on a specific part of a specific protein, so it's unrealistic to expect us to come up with a flawless answer in such a short time, but its not true that just because you see white and gold over black and blue you are not seeing the dress correctly, it is a problem on the neurological level not the primary visual, I bet you would find those who see the gold had higher functioning red cones, and those who see blue have higher firing blue cones, I have asked hundreds of people over the last few days what they see, and the ONLY statistical correlation I can see is that a higher percentage of those who see the blue have blue eyes. My great auntie who has about the worst vision in the world saw black and blue. Anyway it was nice talking! Do you study programming or IT?
deathtokoalas
+Tessa Townsend see, i think you're continuing to express the problem i'm trying to point out. you're looking at the situation as though the picture of the dress isn't really there, as though it's an arbitrary construct of our minds or something, and that it's consequently all entirely subjective. in the end, everybody wins. but, we know the picture is there, and we know what colour it is. so, we have an objective reality to compare our perception to. we can consequently determine whether our perception is correct or incorrect. but, nobody is taking this approach.
i can't answer why people would see the dress differently at different times, but i do think this is where the psychology is important. i think people are easily influenced and would be willing to change their answers depending on peer pressure. so, i have trouble buying the idea that people really see the colour of the dress change. of course, i can't prove that they're lying. but their claims aren't very useful, either. i'd largely consider this inadmissible, or better explained by peer pressure.
but, you're observing behaviour that is clearly associated with vision loss. faded blue is a totally clear marker of vision degeneration.
my academic background was initially in physics, and i've spent time in several programs, but i've completed degrees in both mathematics and computer science.
Adia Aira
+deathtokoalas My interest lies in what could cause a difference in perception... more so than how we perceive the perception... That said, it's all quite fascinating.
I've listened to two explanations: The first pointed to colour constancy,... but reading the definition of colour constancy, i didn't feel like that really explained what was happening. The second pointed to optical illusions... which was my initial assumption, just a trick of the light... but photo editing broke that idea for me. No amount of light-mimicking adjustment would let me see it as white and gold... closest i got was pale blue (where it's blue) and gray (where it's black)... I have seen a few passing colour-blindness remarks, though... so some people are considering simple physical characteristics or limitations...
for the sake of science, +Tessa Townsend i saw black and blue, but my eyes are so dark brown they're almost black. It would be interesting to know whether or not iris colour could play into perception of colour. My eyes are also extremely sensitive to light... Which makes me wonder...
It's not uncommon for humans to wear sunglasses... meaning, after all this time on this planet, many are still sensitive to the bright light that bathes the planet on a regular basis. And i would imagine that the level of sensitivity varies from person to person... In my case, even a cloud day is too bright... so much so that it actually makes my eyes hurt. So i wonder, (and this is just a passing consideration - i'm an artist, not a scientist) could light sensitivity have anything to do with how colours are perceived? Since light, colour, and sight are related...
Tessa Townsend (quintessential)
+Adia Aira I really think that light sensitivity has a huge role in whether your vision corrects the image towards the blue or the red spectrum. But you can be sensitive to red OR blue light as we have different cells that respond to the different wavelengths in our retinas. Thats what I was saying just because you see one colour over the other doesn't mean you are lacking light sensitivity as a whole, it just might mean you have MORE sensitivity to the opposing wavelength (blue you would see blue and black, red you would see yellows and gold). Im exactly the same particularly on an overcast day if Im driving I can't stand how light it is, and I saw white and gold. You sound really smart, great question :) Also as an artist I'm interested, did the colours swap for you at any time? Are you able to see any gold in the image at all? x
Adia Aira
+Tessa Townsend Nope, no gold... just parts where the black is in shadow and parts where the black has light shining on it.
Also, no swapping. I tried to alter the image in the hopes of seeing the colours swap, but all i got was variations of blue and black - from washed out to dark.
deathtokoalas
you'd have to test for it directly, but i wouldn't expect that light sensitivity would be a significant factor in the ability for your eyes and brain to interpret the correct colour, or have any effect on one cone overshooting the other. if one cone is overshooting the other, that's an error inside your eye and it's going to happen regardless of the light source. errors are sometimes accidentally advantageous, but it's just random chance when they are. usually, errors are disadvantages.
i've read some reports suggesting that blue eyes developed in eastern europe as a reaction to sunlight, and indicating that they're an evolutionary advantage. but, if you really look at the way the gene spread, it becomes difficult to connect any cause to effect. and, is ability to adjust to sunlight really going to be that great an advantage? other reports suggest that it may be a result of sexual selection, but i'm not convinced that was ever true. personally, all i can pull out of the data is randomness. there doesn't seem to be any really convincing reason why some people have blue eyes; it seems like there was just a mutation at some point, and that mutation dispersed with no discernible direction or purpose.
again: there's an opportunity to be humble here. our eyes are imperfect instruments, subject to massive amounts of error.
but, you know, it never even crossed my mind that anybody would look at the picture and try and figure out what the colour of the dress actually is in "real life". when somebody shows me a picture of a dress and says "what colour is the dress?", it's pretty obvious that they mean to ask what colour the picture is. otherwise, it's just a dumb question, given the evidence.
and, i don't think that many people meant to answer the question in terms of the object, rather than in terms of it's representation, and am not believing people claiming otherwise. on the other hand, if this really does reduce to a question of what colour the object in the picture is, then it's an exercise in mass stupidity because there's no way to know, and either answer is a guess, unless you're a photography expert.
but, it doesn't really change the situation. first, there's absolutely no way to know what colour the dress actually is from the picture, unless you're a photography expert. anything at all could be done to the picture. i wouldn't even pretend i had the knowledge to be able to undo the effects. and that's kind of the takeaway on it. it's still not a "perception" issue; there's still only one correct answer, and it's still just your brain doing the math incorrectly.
but, i mean, i'd never think to think it. and, if somebody were to really clarify the point that they mean to ask what the dress really looks like if placed in front of me, i'd simply be unable to answer given the sole tool of my eyesight, and make the point clear.
if that's what's actually going on here, what's been uncovered has less to do with lighting and more to do with people being unable to tell the difference between a picture of a thing and the thing itself. it's like mass schizophrenia, or something.
Tessa Townsend (quintessential)
+deathtokoalas It is interesting because our brains often see colours such as light blue and process them as white, because we are so used to photos displaying colours in that way. I think you might be over analysing the psychological elements involved here. A few people have actually said they see light blue and brown, they would be the ones seeing the photo as it is. Everyone else it really is just a trick of the light (ha). And schizophrenia? It's not a hallucination when so many people observe the same thing, reality is only what the majority perceive..
And the greater response to red or blue light in some peoples retina's isn't an error it's just natural variability, the fact that the colour causing the optical illusion can go one way or the other is the thing causing all the confusion. And the optical illusion itself, yes, does point out a slight flaw in our ability to perceive colours in their objective reality.
And it is an advantage in some ways that our eyes trick us, or our brains process blue as white, because snow is white, but in photos its blue, so we'd all be running around saying oh look that snow is blue! But really it's white and our brains know this so they skip a little step in between :P
Schizophrenia symptoms are more like talking to yourself or seeing something no one else sees
deathtokoalas
+Tessa Townsend i kind of think we've stated our differences and are sort of repeating ourselves. you can call it natural variability if you'd like, and i wouldn't disagree with you, but in context i'd define variability as conceptually identical to error. two different labels for the same thing. so, i think, in a sense, you're conceding my point, but you're trying to make it seem otherwise by using different language.
schizophrenia is defined as an inability to determine reality from fantasy. i didn't mean it literally, but it strikes me as the same kind of thing, with people looking into the lighting of the situation interpreting the picture as a portal into a different reality or something rather than as a two dimensional representation of something.
when i look at a picture, i don't determine context from lighting because i know it's a picture. it's two-dimensional. there is no "lighting". whatever is "normal" or "abnormal", that strikes me as a more fundamental difference.
Adia Aira
+deathtokoalas Okay, this is fascinating me for an entirely different reason now. It seems silly to some people, all this over a dress... but i feel like focusing on the object itself is missing what's really happening here.
You're right in that you two are not entirely in disagreement. I even get what you were on about regarding humility. For you, and correct me if i'm wrong, it's about the amount of information you have upon which to base your answer to the question. You do not have enough information to know for an absolute fact what colour the actual physical dress is... so the only question left to answer is what colour the dress appears to be in the photo. It is the only logical thing, in your mind, that they could be asking.
For me, however, the question asked is not the question they may have meant to ask. Usually, but not always, it's easy to determine what they mean to ask... but since i can't know for sure, i can only answer the question they actually ask. What's fascinating me about this is that we're both acting based on how much information we have... for you, it's about the topic at hand (the photo), for me it's about the question.
So when shown a photo and asked: "What colour is this dress?" My mind considers a few things before it even gets to colour: "The only time someone asks me what colour something is, it's to settle an argument, get an additional opinion, or show me an optical illusion." "They didn't ask for specifics." "The photo looks washed out." "It was probably taken inside... i see a picture on the wall..." "This person must be really happy about their new dress." "Are those sleeves?" "What is that? Lace?" With those considerations (amongst others from past experience), some having nothing to do with the colour, my answer is black and blue. Looking at it as "If i have to give an answer..." based on the information i have (gathered from more than just the photo), i feel confident in saying that the dress would definitely be black and blue if shown to me in person.
However, had they asked: "What colour is the image of the dress?" I would have answered "pale blue and charcoal gray" or "pale blue and brown" (depending on the screen settings). Based on the new question, i'm now leaning more towards thinking they're trying to settle a dispute, but i haven't ruled out an optical illusion. Also... this time i feel a need to be specific regarding the colours i see, and i have no interest in the the photo quality or the conditions in which the photo was taken. The dress becomes little more than a palette containing colours to which i'm meant to match names... its real-life colour, irrelevant. It could be a drawing for all i care...
This happens instantly, subconsciously... it's how my mind works. Whereas another person might view those two questions as two ways of asking the same question.
So the colours we see are only part of it. The way our minds interact with existence, our past experiences, our present mood, our relationship with the person asking the question... they also play into how we decide to answer the question... especially when regarding something perceived through senses.
deathtokoalas
+Adia Aira i'm just acknowledging that it's in the realm of possibility that i may have "misunderstood the question", but rejecting the premise offhand and then nonetheless carrying out the "thought experiment" in the situation in which i had.
if we've got millions of people arguing over how to correct an overexposed photo, it's even less about perception. that's literally a math problem.
Zgembo Adislic
but you are the one who perceives that majority, so that doesn't make much sense
deathtokoalas
indeed. it's dumb hippie bullshit.