View Full Version : How does this color illusion work?
Q_Goest
Jun18-06, 07:46 PM
Check out the castle illusion on this page: http://www.ebaumsworld.com/castle-illusion.html
(Instructions on how to view this are just above the picture.)
How do you think this works?
There's another one on this page: http://www.johnsadowski.com/color_illusion_tutorial.html
that explains how to create the illusion. Scroll down to step 9 of the explanation to see the second illusion.
dav2008
Jun18-06, 08:19 PM
I believe this is called an "afterimage"
Basically if you stare at something for a long time and then remove it, you will be left with the image imprint except with the colors inverted.
Edit: Wikipedia has an entry http://en.wikipedia.org/wiki/Afterimage
Q_Goest
Jun19-06, 06:25 AM
Hi dav. I can see the similarities with an afterimage, but I don't think its the same phenomenon exactly. If you stare at the colored picture of the castle, and then stare at a white sheet of paper, nothing happens. You need to actually see the black and white photo for the colors to come out. Also, if you glance away from the dot for even a fraction of a second, the illusion disappears which seems different than an afterimage illusion.
hypnagogue
Jun23-06, 10:13 PM
Illusions of this sort are usually explained by reference to the opponent process (http://en.wikipedia.org/wiki/Opponent_process) theory of color vision:
The opponent process is a colour theory that states that the human visual system interprets information about colour by processing signals from cones in an antagonistic manner. The three types of cones have some overlap in the wavelengths of light to which they respond, so it is more efficient for the visual system to record differences between the responses of cones, rather than each type of cone's individual response. The opponent colour theory suggests that there are three opponent channels: red versus green, blue versus yellow, and black versus white (the latter type is achromatic and detects light-dark variation, or luminance). Responses to one colour of an opponent channel are antagonistic to those to the other colour.
So for instance, in this case, fixating on the dot at the center of the screen helps suppress your natural tendency to make saccades as you view the picture. Because your eyes remain more or less fixed, the cones on your retina picking up e.g. the yellow in the sky become fatigued. According to the opponent process theory, because the receptors detecting the yellow are fatigued, and because yellow and blue are in opposition, reverting to the neutral gray color on the screen causes you to see it as blue.
Here's another example of this type of illusion, but one that doesn't require continued fixation on the greyscale image to work well. Place your mouse cursor in the center of this image (http://wwp.greenwichmeantime.com/time-zone/usa/usa-flag.gif), then fixate on your mouse cursor for 30 seconds or so. After an ample amount of time has passed, redirect your gaze to a white wall or a blank sheet of paper and see what happens.
hypnagogue
Jun23-06, 10:20 PM
Whoops, guess that last paragraph was a bit redundant after checking out dav2008's link. As for why this castle illusion only seems to work with the greyscale replacement-- it may be something about the visual complexity of the image (compare this to the stark colors and simple structure of the flag). Technically, I can see an analogous kind of illusion for this castle image if, after fixation, I look at a blank wall instead of the greyscale castle image-- it's just that, instead of seeing intricate visual details, I just see two blobs of color roughly shaped like the figure / ground defined by the castle and the sky.
Farsight
Sep27-06, 07:03 PM
Apologies if you've seen this link before. In essence it's a colour perception experiment, where the results are astonishing but true. I rather felt it deserved it own thread. Do try it:
http://www.echalk.co.uk/amusements/OpticalIllusions/colourPerception/colourPerception.html
In one picture the central portion is yellow. In an adjacent similar picture the central portion is blue.
Only they're not.
They are in fact both grey.
It demonstrates that colours are not what you think they are. I'm not sure what questions or answers this raises. But it's got me thinking one thing at least: What else do we take for granted?
Moonbear
Sep27-06, 08:58 PM
Farsight, from your description, I've added your post to this thread since it's likely to address your question. However, your link seems to be broken.
Farsight
Sep28-06, 02:45 AM
My embarrassment knows no bounds, Moonbear. I copied and pasted and picked up the abbreviated version. I've fixed it now.
http://www.echalk.co.uk/amusements/OpticalIllusions/colourPerception/colourPerception.html
Rozenwyn
Sep29-06, 12:31 AM
Afterimages happen due to a property of body receptors called "sensory adaptation (http://en.wikipedia.org/wiki/Sensory_adaptation)". hypnagogue almost got it, except that receptors don't get fatigued but they get used to the stimulus. When the stimulus is removed, you get a negative active potential in the receptors, so if you were looking at a blue image, you eyes will see an RGB image of (0,0,-255) which is the equivalent of (255,255,0).
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.