What creates those perceptions, thoughts and feelings?
What are some of the best examples of the uncanny valley?
I’ve noticed that some stylized portrayals can be more appealing than the real thing. In the lion example above, the lion from the Lion King is arguably more appealing than the real lion. Some characters like Wall-E, Baymax, the ponies from MLP: FiM or Elsa and Anna from Frozen are more appealing than perfectly photorealistic characters would be. They don’t just fall in a good spot of the uncanny valley, they’re better than the extreme top right of the curve here: http://upload.wikimedia.org/wikipedia/commons/thumb/f/f0/Mori_Uncanny_Valley.svg/450px-Mori_Uncanny_Valley.svg.png . They are hypercanny, if you will.
I think the opposite of the uncanny valley is probably just the ‘comfort zone’, but it’s a bigger place, and apllies to many more things than humanoid characters, because the uncanny valley is really more of a watershed. A knife-edge where the feeling of wrongness reaches a peak.
Some examples of the opposite of the UV might be things like the 1973 Hovis Bread advert (directed by none other than Ridley Scott). For many people, although perhaps not so much nowadays, it was a little condensed world of normality and comfort. http://www.youtube.com/watch?v=6Mq59ykPnAE
The uncanny valley is generally where something closely resembles a human being in appearance and motion, but is “off” somehow, and produces a feeling of revulsion. The most-popular hypothesis for why we experience this feeling is that we have some sort of “avoid diseased humans” instinct, and nearly-normal-but-not-quite images ping that instinct.
There are basically two options for avoiding the UV: either not trying to make something look realistically human, or nailing it 100%.
But I think there have been a number of studies of this phenomenon in recent years, to find exactly what pings this instinct. So eventually we’ll be able to create robots or whatever that are unrealistic in ways humans don’t mind, or even like, but realistic in the ways that would creep us out if they were otherwise.
Hypercanny is really just a fancy word for “cute”. There are standard cartooning tricks for drawing cute anthropomorphized critters or people. I’ve seen good explanations online but my google-fu is weak today.
In a nutshell, make the critter rounder than real, larger eyes, a smilier face, softer edges. IOW, take a dog-shape, and make a puppy-shape out of it. Or take a 10yo human and make an infantile shape out of it.
Elmer Fudd was cute and lovable because he was really an adult personality doing adult stuff in an infant’s body.
As to why uncanny: It’s almost the equivalent of a joke or an optical illusion. The humor or discomfort or disconnect in those things is that you’re left with two conflicting interpretations.
In the joke, the teller carefully sets up one mental picture in the audience then suddenly at the punch line tells us the actual situation is not at all what we expected. We laugh as our brains grind their gears realiging our internal states to accommodate the new info.
An OK optical illusion is where you see only one thing and when the *a ha *moment hits, you now only see the other thing. A great optical illusion is one where after the *a ha *moment your perception continues to switch back and forth between the two interpretations uncontrollably.
The uncanny valley (crappy name IMO, but not your fault) is the area where your brain is caught in a loop thinking “It’s real; no it’s not; it’s real; no it’s not …” And just like those continuously oscillating optical illusions, they’re entertaining for a few seconds and repellent after a few minutes.
The Lion King animation is very obviously recognizable as “cartoon of a lion”, not “actual lion.” Which makes that comfortable. That cgi almost-real lion in the middle is not obvious and therefore not comfortable.
The fact the Lion King was also drawn for a little extra cuteness / anthropomorphism helps. We recognize anthro-ized animals as a standard cartooning shorthand.
Uncanny is when nearly human looking creations reveal that they are not human. The opposite would be actual humans or indistinguishable representations. Hypercanny if anything might be what we are used to seeing on TV and at the movies, a type of hyperrealism where everything exceeds the average existence. Reality is often portrayed but it’s more common to see overly made up actors, far better lighting and better looking settings than we’d encounter in life.
I think the “uncanny valley” is exactly as Minjin nailed it. I’ve spent time with various weird people, and you know it does not take long usually for the “weirdo-meter” to kick in and tell you this guy is strange, It may be their bearing, demeanour, they way they speak, what they say, how they look at you, etc. We are highly attuned to determine non-verbal signals, and weird or damaged people send off vibes very quickly.
(Ever heard a bunch of women discussing some men? They usually all agree on which ones are creepy" very quickly)
The uncanny valley is that emphasized in all ways. The non-person’s whole body language is just “off”. Some people claim to read a lot in someone’s eyes - but computer simulations or robots don’t put anywhere near the degree of detail into simulating just the eyes. And so on… Humans recgonize patterns, and we see the whole pattern as “off”.
You just may have a fetish. Not judging, I’m just saying…
As for the why of it, it’s that the traits exaggerated are often those deigned to appeal to our built-in child-nurturing instincts - bigger heads, proportionally-larger eyes and smaller mouths. Neoteny, basically - likechibi/SD, a prime example of this sort of thing.-
Even something like Wall-E, his eyes are exaggerated as part of his characterization. Similarly, Baymax has the physique of a toddler - elongated torso in proportion to legs.
And the MLPs are built more like foals than adult horses.
It’s worth noting that Baymax was deliberately designed to be “cute”, even within the context of the movie. The character who created him wanted him to appear nonthreatening and huggable (hence also the inflated body).
And I think that the uncanny lion picture is actually a bad taxidermy job, not CG.
I think a lot of the cartoon characters we’re mentioning here are only “hypercanny” in their context (i.e. a movie). If one of those characters walked out of the screen and stood next to you, I don’t think we’d have the same reaction.
For example, Japanese animation has long gotten away with blue, purple and green hair colors. They seem fairly normal in the context of a cartoon. But when cosplayers try to recreate that? Nope. Doesn’t work anymore. Even the traditional anime big eyes and big heads look freakish when you see it on a mask. What works on paper or on the screen totally fails as something standing next to you.
If “hypercanny” means “realer than real”, I don’t think I’ve ever seen it. But if it means “more appealing than real”, there are lots of examples - in fact, it’s the basis of whole industries. Clothing to exaggerate features: high-heals, corsets, wigs, shoulder-pads… Makeup for skin that’s smoother and brighter than real. Plastic surgery, of course. Body building for bigger-than-real muscles. In art (and photoshop) you can go wild: women with long, long legs and necks, tiny waists, enormous, round breasts; men with super-hero muscles and hyper-angular faces. And, or course, as others have mentioned: hyper-infantile features for extra cuteness.
This reminds me of bird experiments I’ve read about. Some parent birds will abandon their own eggs to sit on fake eggs that are much larger. In one species, chicks peck at a red spot on Mom’s bill to induce regurgitation; if presented with a fake bill with a larger, redder spot, they ignore Mom and go for that. Humans also have attraction triggers and we are similarly subject to hyper stimulation.