Canine Facial Recognition Area Identified

August 5, 2015 7:39 AM

You Seem Familiar Probably not news to us dog owners, but Emory university researchers have confirmed via fMRI that dogs have a specialized brain region for facial recognizance. These are days of miracle and wonder, this is the long distance call.
posted by mrdaneri (31 comments total)

Who's a good dog? Who's a good dog?

YOU ARE!
posted by eriko at 7:51 AM on August 5, 2015 [8 favorites]

SUCH GOOD DOGS!
posted by mrdaneri at 7:54 AM on August 5, 2015 [4 favorites]

Don't you have to sit very still for an MRI? I'm impressed they got the dogs to do that. How do I hire these scientists to train my dog?
posted by Solon and Thanks at 7:55 AM on August 5, 2015 [3 favorites]

Came in to hat tip the Paul Simon reference. Kind of required given my username.
posted by staccato signals of constant information at 7:57 AM on August 5, 2015 [9 favorites]

Yeah, you, do solon I have so many milkbones and so much strongly worded praise for all involved.
posted by mrdaneri at 7:57 AM on August 5, 2015 [2 favorites]

The article says they used a screen to stimulate the dogs, though dogs don't usually react to 2D representations. I wonder why they didn't use real people and objects. I'm sure they could have found a neuroscientist would could use a good interaction with a dog.
Don't cry baby, don't cry.
posted by shenkerism at 7:59 AM on August 5, 2015 [2 favorites]

Real people and objects smell, which would probably throw off the experiment.
posted by griphus at 8:02 AM on August 5, 2015 [3 favorites]

The image search results for "dog MRI" on google reveal the secret sauce of how it is done.

Look at all those good dogs.
posted by idiopath at 8:03 AM on August 5, 2015 [5 favorites]

The trick is to get a lazy dog in the machine, like a pug.
posted by a lungful of dragon at 8:05 AM on August 5, 2015

Cats recognize faces, too, but it also fires off the region of the brain that controls not giving a fuck.
posted by Cool Papa Bell at 8:24 AM on August 5, 2015 [19 favorites]

Very interesting! I have to assume if they did similar scans on other mammals with complex brains they would get similar results (pigs, cats, elephants, or even non-mammals such as crows and parrots, all of which seem to exhibit behaviors of facial recognition as well).

Also, it's hard to believe no one has done the expected "I have no idea how these people got their dogs wedged into their scanners, or why" joke....
posted by aught at 8:32 AM on August 5, 2015 [1 favorite]

I want to assume that, aught, but the lit's light. None of the other species mentioned are our specific obligate partner species like dogs, either. Oh, to be young and and have the patience for grant writing again...
posted by mrdaneri at 8:40 AM on August 5, 2015

i like imagining the frustrated weeping of research scientists as they attempt to coax a cat into an MRI machine, aka the Noise Tube of Suffering. and then hours later just as they give up, having torn at their hair and rended their garments to bits, they look for the cat and it is relaxing motionless in the tube.
posted by poffin boffin at 8:46 AM on August 5, 2015 [18 favorites]

I believe the trick is to get the dog to lay in the machine for a moment, then pretend that you yourself wish to lay in the spot the dog is currently occupying. From my own experimental observations this results in the dog resisting any possible movement from said spot.
posted by borkencode at 8:50 AM on August 5, 2015 [4 favorites]

poffin boffin: seems the kitty burrito method would work here, as long as no tender human flesh-vittles get within fang range
posted by idiopath at 8:58 AM on August 5, 2015 [1 favorite]

Very interesting! I have to assume if they did similar scans on other mammals with complex brains they would get similar results (pigs, cats, elephants, or even non-mammals such as crows and parrots, all of which seem to exhibit behaviors of facial recognition as well).

I swear I once read an article or something about wasps having facial recognition capabilities, either here or somewhere else, but I may be totally making that up. However, I'm not really sure how that experiment would have been done.
posted by litera scripta manet at 9:10 AM on August 5, 2015

One researcher is volunteered to anger the wasps. Everyone then waits for the wasps to be less mad. Then all the researchers go into the wasp room and wait to see who gets attacked.

science is so easy omg
posted by poffin boffin at 9:13 AM on August 5, 2015 [10 favorites]

I'm seeing a fake moustache involved somewhere as a control, poffin boffin.
posted by mrdaneri at 9:15 AM on August 5, 2015 [2 favorites]

I swear I once read an article or something about wasps having facial recognition capabilities

As far I can tell, wasps can only recognize other wasp faces. Crows on the other hand, can recognize human faces and also teach other crows when someone is a threat.
posted by mayonnaises at 9:37 AM on August 5, 2015 [2 favorites]

scanning...scanning...match!
posted by JoeZydeco at 9:50 AM on August 5, 2015 [2 favorites]

I know that one of my dogs has supernatural facial recognition abilities.

A couple of years ago, we had just opened our business (a specialty retail shop.) One day, an unfamilar car came up the driveway and Kenda the Wonder Mutt lost her everloving mind, barking and dancing around, all excited-like. The guy got out of the car and came in the shop and I still hadn't placed him. It was at the end of our conversation that I figured it out: he was a UPS driver who had at some point come to our house and given my dog a cookie.

She recognized him, in a novel context, in his personal vehicle, at a range of about 50 yards.
posted by workerant at 10:03 AM on August 5, 2015 [15 favorites]

I'm excited that one of the quoted researchers, Gregory Berns, is going to be speaking on our science track at Dragon Con next month. He'll talk about dogs recognizing us, how they "love" us, and how they trained their cadre of dogs to stay in the MRI machine.

Afterwards the audience can give Dr. Berns scritchies and feed him a treat.
posted by sgranade at 10:52 AM on August 5, 2015 [2 favorites]

I'm seeing a fake moustache waspstache involved

patent monies now plz kthx
posted by Mr. Bad Example at 10:59 AM on August 5, 2015 [1 favorite]

What's interesting to me is that until recently, I thought dogs really recognized by smell, not sight. Many dogs I grew up with were wary until they got a good sniff of the person ("okay, it's . . . it's YOU!") One of my dogs even barked at me for having my hair in a ponytail.

Then I realized that I grew up with elderly and/or overbred tiny dogs who could barely see.
posted by Countess Elena at 11:19 AM on August 5, 2015

I totally agree, Countess Elena. My family bred and raised labs, and took in rescue dogs for a long time. We had plenty of blind dogs that seemed to work fine with their noses. I'm sure that further investigation will find cross-wiring, e.g smell-noggin' hooks up to the face-noggin'.
posted by mrdaneri at 12:45 PM on August 5, 2015

The image search results for "dog MRI" on google reveal the secret sauce of how it is done.

Look at all those good dogs.

Do not. Its like the Hyperbole and a half dog.

These dogs here look like they'd take a bullet for you. You've got border collies, damned dogs will just -sit there- you train 'em properly. World ending all around them, kept in perfect focus with a raised eyebrow...
posted by Ogre Lawless at 12:47 PM on August 5, 2015 [2 favorites]

Previously (with video!)

And I mentioned in that past post, that this isn't just interesting and adorable neuroscience, it's also some radical animal research methods. The dogs are afforded the ability to voluntarily discontinue the imaging if they feel uncomfortable or for whatever reason; they are not restrained in any way. Being able to stop participating in the research for whatever reason is a literally a legal requirement for human subjects research (45 CFR 46.116(a)(8)), but is... less fundamental in animal research.
posted by Panjandrum at 1:28 PM on August 5, 2015 [5 favorites]

My sister's dog definitely recognizes people by sight. We were walking in the neighborhood and she got really excited when we spotted a young woman with a blonde ponytail some distance away: the hair and her general shape convinced the dog that it was my niece. She had to get quite close to become unconvinced, as well--smell didn't appear to be the issue at all.
posted by suelac at 1:42 PM on August 5, 2015

mayonnaises: "Crows on the other hand, can recognize human faces and also teach other crows when someone is a threat."

omg, you guys, I just had a really good idea about simultaneously training crows all over America to recognize and attack Donald Trump masks, so that everywhere he goes during the campaign, murders of crows swoop down at him and grab for the toupee!
posted by Eyebrows McGee at 4:20 PM on August 5, 2015 [4 favorites]

Remove dogs from tube before energizing.
posted by mrdaneri at 4:56 PM on August 5, 2015

I travel for work and when I am away from home, my dog will search for me during her walks with Mr. Kitty. Apparently, there was a woman getting into a car down our street who looked so much like me, even my husband did a double take for a second - and our dog went NUTS trying to great "me". It wasn't until they walked up next to the car that you could actually see her enthusiasm deflate as she realized it was a random other person. Of course, she then got excited that there was a PERSON RIGHT THERE that she should be able to play with, so the data might be open for interpretation.
posted by Suffocating Kitty at 3:54 AM on August 6, 2015