(Video Link) The rapidly changing faces in this video stimulate your brain to morph them into aliens. As you play the video, defocus your brain and stare only at the "+" in the middle.
The images are all undistorted photos of familiar Hollywood actors. Apparently our visual brain rapidly and automatically tries to resolve the two different faces so that they fit together into a single whole. In so doing, strange exaggerations take place that look like extreme caricatures.
I haven't read a scientist's analysis of this effect, but my own hunch is that the brain is trying to reconcile the right and left visual field of each eye, as well as the images coming stereoscopically from the right and left eye. You can prove this by watching the progression with one eye closed. It still works.
It also strikes me that rather than leveling the differences between the two faces, the brain exaggerates the differences. I wonder, too, if our brain's facial recognition software functions differently on the peripheral retina compared to the way it works around the center of focus. Part of the effect may come from the afterimage of one set of faces switching to another set. Maybe someone who studies this can clear it up for us.
If this effect doesn't work for you, don't worry. Everybody's eyes and brains are different.
-------
Via BoingBoing
Pretty cool! I think you are very close of discovering the secrets of painting.
ReplyDeleteI truly enjoy reading your blog and books.
Kind regards
Oliver
I believe the difference between successive images is really what is driving this, and not the binocular effect. Try covering one side, but still keeping your eye on the cross (since it only works in your peripheral vision). For me the effect is just as strong. The paper seems to confirm my guess:
ReplyDeletehttp://www2.psy.uq.edu.au/~jtangen/TanMurTho2011.pdf
For some reason Sean Penn still looks the same . . .
ReplyDeleteI agree with D. It does seem to work with one face covered. I think it might have something to do with facial recognition and the way our brain recognize faces. Even when we focus on the cross the facial features pop out. I wonder what will happen if there were cars or any still objects instead of faces...
ReplyDeleteYeah, I felt the same thing, that wether it's one face or two the effect is the same...They also become caricatures, so perhaps the caricature artist uses the off focus technique in order to get the exaggeration...You can also use that technique to see better at night.
ReplyDeleteI'm not a scientist but I think I figured this out. All of the faces are tilted slightly to the left. This is especially obvious if you look at the celebrity noses, none of them are vertical. The crosshairs are perfectly vertical/horizonal, which screws up your sense of orientation.
ReplyDeleteI also experience distortion when I look at one of the still images without playing the video, although playing it does exaggerate the distortions further. So the flash face distortion effect D mentioned is probably a factor.
I wonder why this illusion doesn't work on some people. I showed this to one of my friends, and he said all of the faces look normal. He is also more observant than most people, and I wonder if there is a connection between those two things.
Perhaps it's due to the fact that our peripheral vision uses both memory and limited visual cues to "see" what's in the corner of our eye. So the brain "cross-references" what it sees with what it remembers seeing, causing a distortion.
ReplyDeleteJust my best guess. :)
Each of the photos are also carefully aligned with the ones before and after so that the eyes, nose and mouth are in the same relative places. I suspect that similarity causes the brain to exaggerate the other differences outside the facial regions.
ReplyDeleteOur macula which is the size of a pencil point is the only part of our vision that sees detail. Apparently even detail enough to construct an accurate image of a face.
ReplyDeleteSo I think our brain is taking the limited detail that is arriving in our visual cortex and smearing the information. It's filling in the gaps.
I think, our perception of a fully detailed field of view is actually an illusion. Our true detail perception is actually like a flashlight beam in a dark room darting around. We them take all that fragmented information and construct the whole room.
I think this video is demonstrating that phenomena.
Based on what I'm seeing, I think our edge detection neurons are defining the edges of these shards of information. Then our brain repeats pixels to fill in the gaps. But it doesn't repeat pixels in a straight line. I think it repeats them in a Gestalt-like trajectory. The shapes expand in arcs based on movement or memories of similar patterns.
ReplyDeleteAmazind!! O_o
ReplyDeleteAnd all the comments above are very interesting too. I have no personal explanation on this phenomenon… Has the author of this video given any explanation about this effect?
Yes, the exaggeration of the diffs is also what I immediately thought of. It fits with the knowledge that our brain does not capture sensitory info as a whole but only the changes. Cfr Gestalt-psychology.
ReplyDeleteThe fact that we only capture changes in "information" can also be used in the opposite way: to diminish differences. In offices with open cubicles, sometimes a little gadget that produces constant noise is used to cancel out environmental noise. By adding the constant noise, the total noise is averaged out and thus less noticable to our brain.
It's also always a man an a women, maybe the mix of their facial features gives an unsetling result for the brain, something related to gender identification.
ReplyDeleteI always think what evolutionary advantage might this have. One thing, D makes a good point and I was thinking and discovered the same thing, However, if you cover one of the faces as D suggests, look to the side and then with your mouse make the video jump to random points front or back, you still get the same effect regardless which face proceeded it. My guess is your brain is telling us there is something a little off right or left and your brain is forcing you to look there to save you from potential harm.
ReplyDeleteI would say that humans are not adapted to seeing one face instantaneously replaced with another, so we briefly register the two faces as continuous and the same as one another, so we perceive the second face as an extremely strange and unnatural facial expression of initial face, rather than a different face entirely (I think that made sense)
ReplyDeleteHi James...
ReplyDeleteThis "visual phenomenon" is actually more of a mental one in that it has to do with the mind interpreting limited visual information relating to a series of changing but similar shapes & patterns, [i.e:The Face]. And since you are not looking directly at or focusing on each individual face or image & discerning its unique features and their unique proportional relationships & shapes, coupled with the fact the images flashed on screen retain a slight retinal & mental imprint [of that face] which is then slightly transposed over the next set of face images flashed in succession, along with having only an indirect and slightly vague or sketchy perception of the features & shapes in each image means that your brain is obviously doing it's utmost to interpret that limited sketchy [and not directly in your primary field of vision or studied at length] visual information & plausibly filling in the blanks, coupled with the fact that there is a bit of information overload with the dual images being shown in rapid succession either side of your area of visual focus. The faces & their features therefore appear to blend together with the next set and mentally become amalgamations or morphed combinations of a loose visual understanding by your brain with the resultant “perceived images” appearing as warped & grotesque. But the moment you shift your visual point of interest & refocus on the images they are clear & appear to become normal again, but they were never really distorted in the first place. It was all in your minds visually interpreted understanding & building a "plausible" visual context out of what you were only “kinda sorta” looking at.
I get the same effect with one eye closed looking at only one image. It is clear to me that this has to do with peripheral vision. Your brain suggests ways to make sense of things it can not digest fully.
ReplyDeleteAnother example is this picture I made of a combined face of Obama and Romney:
http://www.yjstudios.com/blog/2012/06/05/half-obama-half-romney-for-president-poster/
notice how initially only the Obama face registers because his image is deeper in your brain. To compensate the brain turns the Romney half into Obama.
woops here's link:
ReplyDeletehttp://www.yjstudios.com/blog/2012/06/05/half-obama-half-romney-for-president-poster/
link still doesn't work. just google it if you want to see what i mean.
ReplyDelete