I use optical illusions as a gimmick when I talk to doctors. I do service user involvement with mental health stuff. Clinicians talk about "accepting the reality of the patient experience", but sometimes that concept is quite weak. Showing someone an optical illusion (the grey grid and green bar[1] is great for this) and asking them what colour they see - and then proving that they are wrong - is powerful illustration of how strong the mind can be.
Many years ago I came across a simple graphics program that created caricatures. Details are vague, but it involved plotting the key aspects of a "normal" generic face on in some sort of grid-space, then plotting a more typical (or "interesting" face).
The program then exaggerated the differences between the two. It was fairly basic code, but remarkably effective.
If humans have hard-wiring for facial recognition, then likely a lag between the processing of the facial system and the refresh of other optic-visual processing nodes.
The rods in the eye responding more quickly than the cones is a more simple example which you can see by quickly darting your eyes back and forth in a darkened room while looking towards a red-LED clock.
Why does this effect happen on the computer but not in real life? That is, when a friend (or celebrity) is approaching me in my peripheral vision, why don't I see grotesque person for an instant?
Because usually your friend isn't having their face rapidly swapped out with other faces causing your persistence of vision to perceive average blends.
Both videos (with and without celebs) were pretty trippy.
Maybe it is just me but I don't really see the effect on Halle Berry, Kevin Bacon or Hugh Laurie... they all seems to look pretty much the same to me with or without the face flashing.
Thanks for that, I can see it there, they look like monsters. In the celebrity one I didn't see it at all. The one you link to the faces seem to stay aligned better and more consistent dimensions than with the celebrity version. That may be why it worked better.
It is trippy in a pretty literal sense. I'd wager this is the same mechanism that screws up faces for you when you take psychedelic drugs, except with that you can be staring right at the face and still see it that way. The effect is very similar.
Babies have to learn to interpret human faces so that they ignore the parts that don't matter and focus only on what typically distinguishes them. They manage to do it pretty quickly, but perhaps this is how we see faces for the first few days after we're born.
It would be interesting to get a caricaturists take on this if this is how they see people before they draw them. Also, if they knew of this optical illusion before it was published.
Due to the fact that you are seeing the faces in your peripheral vision for a brief moment, you don't have time to absorb all the detail of the face so your brain fills in the gaps based on memory. Your brain takes shortcuts so the results of 'filling in the gaps' are far from accurate.
That's why you don't even need the second image for this effect. The effect goes away if you focus on the actual image instead (using the mouse pointer as reference point)
Your visual system (well really, all of your sensory inputs) have an effect called adaptation, where they "get bored" of any constant input. Your ears stop telling you about a constant hum, your skin stops telling you about the weight of your clothes, etc. So all the parts of the images that are the same, your optic system gets tired and just stops "seeing" them. So you can literally only see the parts that are different.
my guess is that the brain does something like PCA, but then over-emphasizes the unusual components (probably because that's what makes each face distinct).
in other words, the brain (either hard-wired or learnt) has an "average" face that it uses as a reference. when it has to handle a new face it doesn't "store" the whole face, but only the difference between the new and the average. so what much of your brain deals with is not "the face" but "differences of that particular face from the average".
and somehow (see other replies) this illusion is able to connect that up to your conscious mind without the intermediate step that pads it back out by "adding in" the average.
> Paul Graham's indications for the content of Hacker News:
> What to Submit
> On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity.
> Off-Topic: Most stories about politics, or crime, or sports, unless they're evidence of some interesting new phenomenon. Videos of pratfalls or disasters, or cute animal pictures. If they'd cover it on TV news, it's probably off-topic.
This article is interesting, and it also contains a link to the original paper which is more than many article do.
I use optical illusions as a gimmick when I talk to doctors. I do service user involvement with mental health stuff. Clinicians talk about "accepting the reality of the patient experience", but sometimes that concept is quite weak. Showing someone an optical illusion (the grey grid and green bar[1] is great for this) and asking them what colour they see - and then proving that they are wrong - is powerful illustration of how strong the mind can be.
[1] (http://upload.wikimedia.org/wikipedia/commons/thumb/6/60/Gre...)