h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

"some squares always looked misplaced, no matter where I put them"

Exactly!

I rearranged mine a few times (especially those on the right side) and was ordering them it by whether they were "lighter" or "darker" than their neighbours (*not* by colour) - somehow I hit upon the right combination.

More luck than judgement I suspect!

Protan deficiency is the most common.

I think it's somethign like 8% have a significant (that is to say above the base line varience) protanomolous vision (red-weak).

For Protanopia (red-blind) the proportion in the population is something like 1%

The funny thign about the smiley is that it isn't grey; I can see that it's banded and since it's a spectrum can guess it's red through yellow to blue at the bottom.

BUT the hair, isn't really red ..it's more than kind of grey which means it's probably red but I'm not seeing it.

The yellow and green, one's slightly darker than the other but I'm not sure I'd identify the green as green in another context probably more likely to say orange.

The entirety of his chin is blue, which probably means at least half of it is purple of some description or possibly pink.

This is the phenomenon of what I mean by "guessing" I can see a hue but it's really a kind of algorithm of context and similarity and judging by lightness and darkness, that I get by on. What I'm not able to do is tell colours apart by the hue.

Just by way of a check - I tried it once more but his time I didn't bother trying to reorder them by "Brightness" once I placed them; I just went with first-instinct (sorta like Arthur Dent plucking chess pieces out of a bag: "WHAT DO YOU GET IF YOU MULTIPLY SIX BY NINE".")


http://images112.fotki.com/v189/fileJQkn/7c8b9/2/489822/7537268/Picture4.png
http://images19.fotki.com/v285/fileyR9c/7c8b9/2/489822/7537268/Picture5.png

It's kind of a puzzle that my results aren't more consistent with *each other* - I think that's probably down to the fact that when I say I am guessing I really am guessing (!) but I am consistently lining up with the the protan deficiency and each time towards the strong-to- severe end of the gradient and not the other way.

Yes, I tend to go by brightness in these tests too Hence my initial confusion in the other test you linked to (the one where you can test for the protanopia, deuteranopia and tritanopia separately), because they asked to pick a colour which was "close" to the one above, but when I see three dark colours I don't find them close to a light one

And I don't see the grey either (I see basically orange - yellow - blue, but would possibly detect more on a bigger model). Just trying to confuse the so-called "normal" people.

Couldn't (shouldn't?) we coin some derogatory term for them?

Well, with this test I can more or less arrange the squares in three big regions (with the odd square not fitting anywhere ). I guess this is something I woud consistently do, while within each region the squares would be more or less randomly arranged. This could be an indicator of something they take into account?

(Maybe I will retake the test. I always hope not to get it all right by pure chance though!)

I tend to go by brightness in these tests too - I think that's something few people understand about colourblindness is that often the deduction is made by relative brightness. To be honest I didn't realise that's what *I* did until it was pointed out to me: I thought I was trying to tell colours apart - which is impossible if you can't *see* them - now I realise what I think of as different colours are typically too similar hues to tell apart but one is brighter than the other.

That's the essence of the flavour-blindness meatphor I was trying to describe initially, brightness of a colour is analogous to the intensity of flavour being the difference between a weak cheddar versus a powerful brie. you can't tell it apart from tofu but you can sense the intensity which lets you think "ah this one is probably cheddar" oh it's emmantal - well I was close!

"I can more or less arrange the squares in three big regions (with the odd square not fitting anywhere)"

Yeah I know what you mean the bluish/pinkish ones to the left the brownish ones to the right.

"within each region the squares would be more or less randomly arranged. "

That where the light/dark sorting would come in "I might shuffle than one over one because it's slightly lighter" I'm thinking but that's not telling them apart by what colour they are, that's something else entirely.

But this kind of sort and shuffle is atypical behaviour - most people just recognise what colours there are, for people with colourblindness it's the whole time-consuming cognitive process of sorting and matching or else a default - oh bugger I can't see that, what colour is it supposed to be?

"The other test you linked to (the one where you can test for the protanopia, deuteranopia and tritanopia separately), they asked to pick a colour which was "close" to the one above, but when I see three dark colours I don't find them close to a light one"

I know. Devilish isn't it?

It's set up that way to eliminate brightness as a guide forcing you to go by hue. For anyone with a colour vision deficiency this is next to impossible, the error rate goes up and the diagnosis is evident.

Just to explain the mechanics of that one though it something like this.



Big paint splat, let's say a fetching shade of purple with a brightness of, say 36.



1 is a pseudoisochromatic blue with a brightness of 5
2 is the same purple with a brightness of 5
3 is a pink with a brightness of 5

If you are in a position of confusing your purples, blues and pinks and relying o brightness to distinguish them, matching the paint splat to the correct option is going to be very tough.

Anyone without a colour vision deficiency will acknowledge the difficulty (the options being of altered brightness make it pretty tough for anyone to say which is the right match, but if they can tell hues apart, determining which is the purple from amongst the blue and pink will be relatively swift ; where as you and I will puzzle over it and probably end up guessing.

Since Amy P suggested it - I've written up my original journal with a touch more elaboration on a few points and I'll be submitting it to the AWW for inclusion in The UnderGuide.

A63242679

Just a quickie, Clive!

The smilie actually goes, from top to bottom, red, yellow then a touch of purple at the bottom....



GT

Cheers GT.

Anyone wishing to comment on the (slightly longer) Article in AWW then the thread to do so is here:

F74130?thread=7300670

Top is red. I think there's a very thin orange band, but it's hard to make out. Then yellow. That's easy to see. Then a very thin green band. It's thin, and seems to just at both edges of the smiley, missing the middle, but green is a distinctive colour and I can see it clearly. There are then two bluish shades at the bottom: they're close colours, so the dividing line isn't distinct, but I can see it if I look closely. Indigo and violet, I'd say. There's no actual "blue", as such. Certainly royal blue is not present. I've always found it hard to put proper names to the colours at that end of the spectrum.

Who made that smiley?

TRiG.

There's orange behind the red eyes, but none on the edges. I *think* I remember hearing that Amy the made it.

I just did this one http://www.biyee.net/v/color_vision_test/index.htm

and these are my results

Test Protanopia Deuteranopia Tritanopia
Adjusted Score (%) 100 96 97
Test Time (UTC) 2010-02-14 22:06:56
Total Test Number 42 43 57
Correct Answer Number 42 42 56
Wrong Answer Number 0 1 1
Correct Answer Portion (%) 100 97 98
Average Response Time(s) 1.82 1.83 1.66
Response Time Stdev(s) .94 .7 .75

WEll according to this:

"

All of the reported data are simple, straight forward and self-explanatory statistical parameters except the adjusted score which is the most important number.

Adjusted score, in a nutshell, is the estimated true portion of correct response in terms of percentage. For example, an adjusted score of 80 means it is estimated that 80% of the time the testee truly knows the correct color and the rest (20%) of the time he just guesses. Since there are three choices, a completely color blind person can get 1/3 of his responses right by randomly picking the choices. Therefore correct portion of 33% is translated to an adjusted score of 0.

It is possible to have negative adjusted score. This means the testee's color vision is even worse than complete color blindness. In other words, the wrong colors are more likely to be picked than the correct ones. In reality, the negative adjusted score is usually a result of statistical variation, especially of small sample numbers. The design of this test, due to its uniform image pattern, uniform background and random brightness, does not allow any inherent negative adjusted scores."


I'd say you did well.

I just went up to "very good" on the protanopia test and got an adjusted value of 32%, which if I read the above correctly means roughly a third of my 67 answers were accurate and two-thirds were guessing.