One of the hardest things about doing a behavioral experiment on an animal is to not anthropomorphize. Meaning, when an animal is doing something, you can’t assume anything about its behavior is remotely human. Example: say you’re studying squirrels, and squirrel one chases squirrel two off a branch. You cannot then say squirrel one doesn’t like squirrel two, or squirrel one was mad at squirrel two, all you can say is squirrel one chased squirrel two off a branch. Unfortunately, anthropomorphizing is super easy to do, especially without even realizing you’re doing it. I do it all the time with my dog Lou, sometimes for fun (like the ongoing and conflicted relationship saga he has with my friend’s beautiful border collie/aussie mix), but also sometimes completely accidentally. It’s the way some people trick themselves into thinking their pet snake loves them (sorry reptile lovers, but your scaly friend just doesn’t have an “i love you” brain), and it can be a major issue for anyone doing behavioral research.
I know, because I love behavioral research. I think behavior, human and animal alike, is one of the most fascinating topics out there. And I know first hand how easy it can be to want to believe an animal is able to feel something so bad you make yourself believe it’s true. Need an example of how this can affect research? Oh how handy, I just happen to have one in front of me!
Yesterday I read an article on Slate titled “A Parrot Passes the Marshmallow Test.” For anyone who isn’t super into behavioral science, that probably sounds like the weirdest clickbait ever. And it kind of is. But it does actually make sense if you know the back story. The marshmallow test was created by Stanford professor (woo Stanford!) Walter Mischel in the 70s and it basically goes like this: put a kid in a room with a marshmallow, tell the kid to wait until you come back and if they haven’t eaten the marshmallow by the time you get back, they get two marshmallows. Pretty simple yeah? Well, it’s apparently pretty hard if you’re seven. I’ve seen video of people doing this experiment (with gummy bears instead of marshmallows) and it’s hilarious. Kids talking themselves down from eating the candy, squirming in the chair, one loophole finder even licked the candy then put it back. Besides enjoying watching kids squirm, it turns out this way of testing self-control is a pretty good indicator of a lot of things down the road, like higher SAT scores. So if you were wondering yes, I will definitely be administering this test to my future children.
Anyway, as you can probably now guess from the title, a research team at Harvard University claims to have given this test to a parrot and had it successfully pass the test (they used other treats instead of marshmallows). Now unfortunately these findings haven’t been published as far as I know, but were presented at the Animal Behavior Society Meeting last month. Since I was not at said meeting I cannot speak with as much expertise on the subject as I would like. But going off of what the Slate article says, the researchers gave the parrot a cup of food, told it to wait, and after 10-15 minutes came back with a cup of even better food for the parrot, which he got if he didn’t eat the first cup of food. Now I don’t know about the scientists in charge of the study, but to me this just sounds like they taught the parrot a command and didn’t actually test his impulse control.
Lou knows how to wait. I can put treats on both his paws and tell him to wait, and like a good little doggie he will not eat those treats until I tell him it’s OK. But hold on a second, I say to myself as I type this, me being in the room with him and having him wait a few seconds is so not the same as what that bird did! And I was right, so I decided to do the experiment with him. I put him in the kitchen, put a treat on the floor, told him to wait (admittedly I did not explain he would get two treats if he did wait, so maybe that was unfair) and locked him in there for ten minutes. And when I walked back in what did I find? Treat on floor, Lou in exact same place I left him, a look on his face saying “why do you torture me so!” (Hint: that was anthropomorphizing). Hooray! My dog will get a 1600 on his SATs!
OK, so other than bragging about how well trained my dog is (errr..sometimes) what am I really saying? That it can be incredibly easy to turn your wishful thinking into misinterpreted results. Now, I would be totally jazzed if this does get published and I find out that there was a whole chunk to this story I’m missing, but I really seriously doubt it. How do you explain to a bird that in ten minutes he will get another treat but only if he doesn’t eat the treat you’re giving him now? Unless we have suddenly come up with a universal bird/human language, you can’t. So as fun as it is to read stories about animals doing something amazingly human, take them with a grain of salt (or if you’re patient, two marshmallows).