Interesting study on facial recognition that applies to our discussions...

Collector Freaks Forum

Help Support Collector Freaks Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Michael Crawford

Super Freak
Joined
Sep 11, 2005
Messages
5,233
Reaction score
45
We were discussing people and their views on portraits on a recent HotToyCast, and this study came out today that is very relevant. Yes, you have to listen to this one, but it's worth your time if you're interested in the wildly various opinions on sixth scale portraits. In a nut shell: for most of us, outside of the people like friends and family we see every day, we're really ****** at recognizing faces. But that's only half the story. Perhaps more importantly for collectors, while most of us are really bad at it, we think we're really good.

So next time someone says a sculpt looks just like the intended person, or they say it's terrible ("that looks more like Broderick Crawford than Joan Crawford!"), remember that a) their most likely really terrible at recognizing faces and b) you are too, so it's pretty much a wash.

Deciding whether a portrait looks like the actual person is far more subjective than even some of us thought, and is more like a function of art than quality.

BTW, if anyone comes across a link to the actual study, I'd love to see it. I think it applies even more to our situation (recognizing) than the title implies (remembering) because he talks about how they used photos from different angles and situations to see if people could compare two faces. Pretty much what we do with our toys all the time!

https://www.npr.org/2017/10/27/560346433/were-not-as-good-at-remembering-faces-as-we-think-we-are
 
Being able to recognize faces is a crucial part of life. Some of us are very good or bad at it, but in general we aren't as good as we think we are.
that's all i see. no further text. what do i do wrong?
 
Backs up what I have been saying for a long time. I always compare sculpts to several pictures of the subject and don't rely on memory. It still amazes me how the importance for many in this hobby has shifted from accurate likeness to detail and paint apps being more significant. I don't care how real a head looks, for me it has to look like the person it is suppose to represent. So many figures being praised for high detail when the likeness is mediocre at best. To each their own. Despite the nitpicks it is a great time to be a 1/6th scale collector.
 
You have to press play in the upper, left-hand corner.
oh i see. i prefer reading.
p.s. today they added the text!
i myself have a strange recognition mechanism. when i see a familiar actor in a movie, i can easily not recognize him/her (especialy with changed haircut) until he/she starts talking - at that moment i recognize either the voice or the way face muscles move, or both. also noticed there aren't many patterns face moves, but they are all very distinct from each other.
so with our "toys", having still faces forever, it's difficult to sculpt a face that i will call 100% accurate, because it doesn't have a benefit of moving and thus showing me "yes, this is that person".
probably the reason why i stick to robots and aliens in my collection :slap both are either stone faces or rubber heads with not many natural expressions in the movie, so it's not that difficult to replicate them.
 
oh i see. i prefer reading.
p.s. today they added the text!
i myself have a strange recognition mechanism. when i see a familiar actor in a movie, i can easily not recognize him/her (especialy with changed haircut) until he/she starts talking - at that moment i recognize either the voice or the way face muscles move, or both. also noticed there aren't many patterns face moves, but they are all very distinct from each other.
so with our "toys", having still faces forever, it's difficult to sculpt a face that i will call 100% accurate, because it doesn't have a benefit of moving and thus showing me "yes, this is that person".
probably the reason why i stick to robots and aliens in my collection :slap both are either stone faces or rubber heads with not many natural expressions in the movie, so it's not that difficult to replicate them.

That sounds to me somewhat like facial blindness:
https://en.wikipedia.org/wiki/Prosopagnosia
 
:lol
i believe though that it's a developed mechanism caused by my myopia that had to be surgically stopped in early childhood but had already reduced my vision to -4. and my refusal to wear glasses for many years.
i learned to rely on voice/steps/movement/silhouette/face expressions combination rather than on still faces themselves, as i didn't see those faces anyway.
 
Star Wars is prolly the one trilogy we all know by heart and Harrison Ford has been in many more movies since, so we recognise every expression of his from every angle, while a head sculpt is only a single expression we can see at every angle. Also digital cameras take pictures at finer resolutions than our eyes can make out. The detail you see in all these photos we take is like using a magnifying glass. Technically it is.

Anyone ever put up scaled up pics of each headsculpt at 1:1/4/6/12/18 to compare?
 
If you enlarge some of these detailed sculpts to life size the texture looks like beat up leather or the surface of a golf ball, not human skin. It tricks the eye into thinking it looks real at 1/6th scale, but in 1:1 reality not so much.
 
If you enlarge some of these detailed sculpts to life size the texture looks like beat up leather or the surface of a golf ball, not human skin. It tricks the eye into thinking it looks real at 1/6th scale, but in 1:1 reality not so much.

of course at 1/6 scale you only get 1/6 details, same as jpg, try displaying a 720p photo on a 4k display, what looks good on a 720p monitor or 1080p may not look good on a 4k display, but i do understand 4k tvs has great upscale function but still lots of details are lost.

that is why most people shoot movies, photos and prototypes in bigger scale so they can compress and resize to smaller scale during production.
 
of course at 1/6 scale you only get 1/6 details, same as jpg, try displaying a 720p photo on a 4k display, what looks good on a 720p monitor or 1080p may not look good on a 4k display, but i do understand 4k tvs has great upscale function but still lots of details are lost.

that is why most people shoot movies, photos and prototypes in bigger scale so they can compress and resize to smaller scale during production.

What I was saying is that the texture a lot of people think looks so real isn't the right scale. It tricks the eye into thinking it looks real until put into proper context, than it doesn't look real at all. Lots of collectors in this hobby giving mediocre likeness work a pass because OH THE DETAIL. They think it looks like a real person so it must be a great sculpt even if the likeness isn't nailed. If that is what people like, that is all good. Personally if I am going to buy a figure of a tv or movie character I like I prefer it actually look like the person it is suppose to represent.
 
We were discussing people and their views on portraits on a recent HotToyCast, and this study came out today that is very relevant. Yes, you have to listen to this one, but it's worth your time if you're interested in the wildly various opinions on sixth scale portraits. In a nut shell: for most of us, outside of the people like friends and family we see every day, we're really ****** at recognizing faces. But that's only half the story. Perhaps more importantly for collectors, while most of us are really bad at it, we think we're really good.

So next time someone says a sculpt looks just like the intended person, or they say it's terrible ("that looks more like Broderick Crawford than Joan Crawford!"), remember that a) their most likely really terrible at recognizing faces and b) you are too, so it's pretty much a wash.

Deciding whether a portrait looks like the actual person is far more subjective than even some of us thought, and is more like a function of art than quality.

BTW, if anyone comes across a link to the actual study, I'd love to see it. I think it applies even more to our situation (recognizing) than the title implies (remembering) because he talks about how they used photos from different angles and situations to see if people could compare two faces. Pretty much what we do with our toys all the time!

https://www.npr.org/2017/10/27/560346433/were-not-as-good-at-remembering-faces-as-we-think-we-are

Comparing facial recognition of an aspirational, frequently seen icon like for example Harrison Ford to a person you briefly worked with a decade ago (that you might struggle to recognize) isn't really apples-to-apples.
 
Back
Top