The ability to recognize somebody’s face is an extremely important ability that humans have developed over the years. I mean, think about it. What’s more important than being able to distinguish between your wife and a homicidal maniac lying in your bed? But just on the tails of recognition’s importance is the ability to discern emotional states. Again, nothing is much more important than discerning between when your wife is going to kiss you versus bite your nose off.
The question arises, however, are these two abilities one in the same, or do they rely on separately evolved parts of the brain? And either way, are these parts of the brain unique to primates, or specifically humans? There’s a whole body of research on the subject. The initial thought was that the two abilities were completely separate, partly because early experiments showed that people with brain damage who couldn’t remember a face for the life of them could still discern different emotional states.
However, later studies have shown that it is impossible for people to process an emotional expression completely independently of identity. It’s a weird asymmetry. You can pick up identity and ignore emotion but you can’t pick up emotion and ignore identity.
So again, the question is, why is that and is it just a human thing?
To find out, Ed Wasserman from the University of Iowa trained pigeons to recognize both emotional status and identity. Why pigeons? They’ve got killer eyesight and yet are on a completely different branch of the evolutionary tree. If they show the same asymmetry, then it’s quite probably that the brain functions that allow humans to recognize faces aren’t uniquely human. At least, not entirely.
And if you think pigeons can’t tell people apart, think again. I mean after all, it’s very important for you to be able to differentiate a happy dog from a rabid dog, so it stand to reason that other animals can tell angry from sad in other species as well.
In a preliminary experiment, it was shown that it was indeed possible to teach pigeons to recognize faces that share identity and emotion. Then, in the second study, the researchers tried to see if the birds could separate the two. In trying to categorize photos based on only either identity or emotion, the birds found it easier to group identity together while ignoring emotion than to group emotional faces together while ignoring identity.
How do you train pigeons, you might ask? The birds were put in front of a computer screen that displayed four images at a time, each in a separate corner. They also were presented with four buttons in four corners that corresponded with each. After learning that pecking the screen resulted in food, the trials began. One group had to match the face that appeared in the center of the screen with the same person in one of the four corners, while the other group did the same with emotions. If the correct button was pushed, food was given. If the incorrect button was pushed, the screen darkened and the trial went into “timeout” for about 30 seconds, and the bird was given another shot at the same image. This continued until the pigeon got the right answer, though it only counted as “correct” if it got it right on the first try.
The researchers did this 160 times each day and eventually the birds got it. They were considered “trained” once they hit a criterion of 74% correct responses overall with at least 55% correct responses on each of the four buttons in a single session.
Some birds got the hang of it in just 14 sessions with others took 74 (dumb asses). But the important number is the average. It took the “identity” identifiers an average of 37 sessions to get it right while it took the “emotional” identifiers an average of 89 sessions. That’s a big difference.
What it means is that there is evidence that birds share the same sort of facial recognition asymmetry that humans do, lending credence to the theory that we have a little bit in common in brain function when it comes to that area. So chances are that whatever regions of the brain are recruited for facial recognition, they developed the ability long before humans and birds split evolutionary ways.
In short, the ability to recognize faces and emotions may not be as human as previously thought.