Elephants Have Self-Concept

The Proceedings of the National Academy of Sciences will be releasing a study soon that shows elephants can recognize themselves in a mirror.

Among animal psychologists this is taken as evidence that they have a sense of self, a self-concept. Researchers gave three captive elephants the mirror and mark tests that were invented in the 70s by Gordon Gallup, a psychologist. Animals who don't recognize themselves in the mirror either ignore it or attack it. Those who do understand they are looking at themselves often inspect parts of their bodies that they can't normally see.

Chimps check out their behinds, the inside of their mouths, and their eyes. Elephants evidently do the same thing. The mark test is a variation on the mirror test. Visible and invisible paint is placed on either side of the individuals head and then observers watch to see whether the painted animal touches the visible paint spot but not the invisible one.

Members of all great ape species (chimps, bonobos, gorillas, orangutans) have passed these tests, but not all individuals do. Infants and juveniles don't "get it" and neither do infant humans. There is some developmental process that has to happen first, but at some point in time, usually by their first birthday, all normal humans understand the mirror image. Adult dolphins are the only other non-ape to pass these tests — that is up until now.

What's really amazing is that through evolution by natural and/or sexual selection, three very different kinds of critters all wound up with an ability to understand themselves. I wonder if this means they are conscious beings. Do they think about themselves? What do they think?

A while ago when Marc Hauser, a monkey psychologist who has done research on captive marmosets, released his book *** Wild Minds (excellent!!) the NY Times ran an article with a sidebar on testing your dog for self-concept. Max flunked it miserably. As far as I can tell he thinks mirrors are utterly boring. When we got our new pup I was keen to observe her in front of mirrors. We hadn't hung one of ours yet so it was on floor level with her. She was very, very interested in it but didn't do anything that confirmed in my mind that she knew she was looking at herself. I think she may have thought it was one of her litter mates. When we walked her around campus the first few times, she'd spot her reflection in the windows of doors. I think her reactions to them speak more of her outgoing, interested-in-other-dogs personality than her self-concept. Perhaps she does know the image is her, perhaps not. Either way, it's still fun to watch her watch herself. For that matter, it's also fun to watch her watch TV. She really loves dogs and balls. Max never pays attention to TV. Because of these differences I find it hard not to think that Max's new pal is smarter than he is. I think we humans assume that anytime another animal does something human-like that means it's "smart" even if it's a dumb as watching TV.

Some headlines on the elephant self-concept story reflect this assumption.

*** Adult captive cotton-top marmosets have passed these tests according to Hauser's research. He used day-glo neon pink Manic Panic hair dye to color the marmosets' stark white mohawks. This got them to look in the mirrors long enough to realize themselves staring back. Monkeys perceive direct eye contact as a threat (as do dogs) so their instinctual response is to retreat or attack. If they could just get over the eye contact part of it, it's possible that species who respond to eye contact this way would be able to "pass" the tests.