According to a recent study by Kim Curby and Isabel Gautier of Vanderbilt University, we can remember faces far better than, say, wristwatches or cars.
“Our results show that we can store more faces than other objects in our visual short-term memory,” says Gauthier. “We believe this happens because of the special way in which faces are encoded… Being able to store more faces in VSTM may be very useful in complex social situations.”Addendum May 17
I was intrigued by Erik's comment about hair and beards, and noticed the lack of them in the pictures that were used for the experiment, so I asked Dr. Gautier the following: "Have you factored in the role that hair or beards play in facial recognition? Does it code or store differently from other features given that it can be more variable?"
She replied: "In general, our work and that of others focuses on mechanisms that appear to distinguish face processing - the internal features of faces are processed in a more holistic manner (not as parts, but as a whole) than other objects. In the context of experiments where we only use a limited number of faces, hair might be SO helpful that people would rely on it entirely and we could not study how internal features are processed - so they are generally excluded."
Vanderbilt University report, link and article in Science Daily, link.
12 comments:
This subject of recognizing faces has facinated me as well. How we all have the same basic elements that make up a face, placed within a few centimeters of every other face, and yet each is so recognizable. Even when you run into an old class mate that you havn't seen in twenty years and they've put on ten pounds and gravity has moved those features a little south. It must be a survival trait to recognize friend from foe.
Great sketch as always.
Larry
Fascinates me too !
For comic books it's a well known fact that if your characters have very specific features (moustache, beard, very big or small nose) it's easier to drawn them recognizable every time.
Also, old characters are much easier to draw in a recognizable way than young people, since young people have too 'standard' faces and thus the features you would have to draw would be very subtle.
So, don't ever try to do a comic book about the 10 candidates of the 'Ohio Miss Blond Elections' or something! You would have to be very creative to make those 10 smouth-faced young blond characters distinguishable!
The thing is, you can't trick the viewer : I noticed recently that in a very small picture on the internet, without context to what the picture was about I could recognize a certain person.
Her face was just a few pixels high! And there were no important hints in the image to help me (I think).
Another proof of how good we are: line me up a German, French and Dutch person and I would probably guess their nationality.
Azians can do this to (while for us Caucasians they 'all look the same' don't they?).
I wonder, could US people guess what state someone's from? Or see the difference between an American and a Canandian person?
Hey, even sheep (don't they really all look the same?) can recognize all their individual gang members!
Bottom line: we are scaringly good at this, so watch out you draftsmen!
Scarily good is right. I stand in awe of artists who can do caricature well. They stretch features like taffy and still the person is recognizable. I wish I understood how it works. (I've never seen anyone even try to explain it. It must be intuitive, right?) Seems like I'm missing the appropriate brain area, because I struggle and struggle, and I can almost never get a likeness that satisfies me.
If there is a secret, someone please tell me! :-)
The key to caricatures is extreme familiarity with the features of the human face and lots and lots of practice. The human brain can recognize a face from very little visual information (a pair of dots and a curved line, for instance), and a caricature sees how far that recognition can go before a face becomes too distorted. It also helps if the artist has an irreverent sense of humor and an active imagination.
Ah, that pesky practice thing.
It's encouraging, though, because it suggests I still might have a chance to learn how to do it.
Have you seen the work of Paul Ekman on facial expressions? That's what this reminded me of.
I wanted to take your oil paint poll, but the brand I usually use--Utrecht--isn't on there! :)
I just found your blog and I love it! I was reading through and I found this post.
And recognizing people has always been hard for me. I just found out a year ago I'm almost completely face blind, and my art teacher thinks that's how I can paint faces as well as I do. Faces are nothing but shapes to me.
My facial VSTM is broken.
So, while it helps me with as an artist, I've offended countless people. Including my own mother. But it's all right. I used to think, wow, I wish I could recognize people, but now I wouldn't trade my ability to recognize them for my ability to draw them. :P
Ally, thanks for your comment on this and other posts. The face-blind question is a fascinating one. I think every artist has to "turn off" the facial recognition engine to see the pure shapes in order to paint a portrait, but I feel like you have to turn it back on again for fine tuning and for caricature. Sometimes I look closely at the features of a face and wonder what makes that nose or that eye belong to just that person.
Quickly becoming my favorite blog. Thanks so much! Colin and James, here's some info on how we can recognize caricatures.
On Intelligence
Copyright © 2004 by Jeff Hawkins and Sandra Blakeslee
70
four attributes of neocortical memory that are fundamentally different from computer memory:
• The neocortex stores sequences of patterns
• The neocortex recalls patterns auto-associatively
• The neocortex stores patterns in an invariant form
• The neocortex stores patterns in a hierarchy
80
let's return to the sensory cortex and look at music again. (I like music as an example because it is easy to see all the issues the neocortex must solve.) Invariant representation in music is illustrated by your ability to recognize a melody in any key.
Think of the song "Somewhere over the Rainbow." You probably can't recall the key she sang it in (A flat). If I sit down at a piano and start to play the song in a key in which you've never heard it — say, in D — it will sound like the same song.
The memory must store the important relationships in the song, not the actual notes. In this case, the important relationships are the relative pitch if the notes, or "intervals." "Somewhere over the Rainbow" begins with an octave up. followed by a halftone down, followed by a major third down, and so on.
81
Similarly, the memory of your friend's face must also be stored in a form that is independent of any particular view. What makes her face recognizable are its relative dimensions, relative colors, and relative proportions, not how it appeared one instant last Tuesday at lunch. There are "spatial intervals" between the features of her face just as there are "pitch intervals" between the notes of a song.
When you memorize her face, you memorize these relative attributes.
Michael, I'm glad you're enjoying the blog, and thanks for sharing these insights into how we store memories. I'll look into that book.
Extending then from familiar melodies to the holistic grasp we have of faces, it's amazing how we can recognize a face not only at different angles, but in different lighting, with beards or sunglasses, or covered with face paint.
And it's interesting how we can recognize a face and fail to remember the context or the name, as if that is coded differently.
nice
Post a Comment