Avian Expert Articles

Pepperberg’s Parrots Surprise With Word Choices

“Hidden” Words

Griffin being trained on “3-corner wood.”

My parrots are trained to communicate with humans using English speech, and most of the time they do pretty well. Not only do they respond to our questions asking them to label colors, shapes, materials, and toys, but they also ask us for various items or to be moved to different locations. So it’s quite common to hear Griffin saying something like “Wanna nut” (a treat available only during training or testing sessions) and “Wanna go back” (to his cage), or Athena trying to produce a clear version of “chair” so that she can join her human companions on the armchairs in the lab.

Sometimes, however, the birds simply stop producing certain vocalizations; these labels just seem to disappear. We think it is because the items to which the labels refer (usually healthy foods or familiar toys) are always available, and as a consequence the parrots do not often need to request these things. Alex kept most of his labels active because he would frequently comment on what he was eating or chewing apart (e.g., respectively, “yellow corn” or “wood”), even when we didn’t require him to do so. However, neither Griffin nor Athena seem to think that kind of chatter is important and don’t engage in such behavior. But we’ve learned that labels that aren’t commonly used haven’t entirely gone away; rather, they are “hidden.”

Proof That Parrots Do Not Forget Words

For example: We recently had a film crew in the lab, and thus the parrots’ lunch was not on time. Too, their breakfast food bowls were not in the usual spot atop their cages because of a particular shot that the crew wanted. All of a sudden (and, of course, after the camera was turned off!), we heard Griffin’s very plaintiff “want banerry!” (Alex’s and Griffin’s label for apple). One of my students, who had been in the lab for over a year, was totally taken aback…this was the first time she had heard the label!

We sometimes joke about Griffin’s roughly once-a-year request for “grain” (cooked grains), because this food appears like clockwork twice a day and only rarely does he decide that he wants it at an unusual time. We even saw a related behavior with Alex. He really liked fresh cherries that we pitted for him, but they are a seasonal fruit and thus were rarely available. However, each year, as soon as we could access a new crop from the grocery store, he quickly identified this favored food, using a label that he hadn’t uttered for many months.

Parrot Surprise Words

Other times, we have no idea that a bird has learned a label until it is used in a novel situation. One day, the students forgot to replace Athena’s water bowl after the noon-time cleaning of her cage. About an hour later, we heard her saying something that sounded very much like “wa-er”… unlike anything she had previously uttered … and realized our mistake. We wouldn’t replicate the circumstances by choice, but do wonder if it would happen again.

And then there was the time that we were trying very hard to teach Alex “seven”… Saying “v” without lips is difficult, and training was going slowly; he was approximating the label by saying “s…none” or “s…one” (Pepperberg, 2007). Everyone in the lab was quite surprised one day to hear that vocalization coming from a different cage…Griffin had clearly been attending to Alex’s training and decided he could get some attention if he produced this approximation…and, of course, he was correct.

Parrots Getting Words Right

Finally, we know from an earlier study (Pepperberg, Brese, & Harris, 1991), that Alex would sometimes practice a label in our absence until he got it right…we learned, but only after we set up a tape-recorder to track his behavior when he was alone, that he was actually trying to say a new label weeks before he produced it clearly in our presence. We still aren’t sure why he engaged in such behavior, but maybe he just wanted to avoid the somewhat negative feedback (“Alex, you’re close…say better!”) he received when he practiced during training.

The take-home message is that even with birds that live in a laboratory and are being trained to talk — we never know quite what we will — or will not—hear!

Pepperberg, I.M. (2007a). Grey parrots do not always ‘parrot’: Phonological awareness and the creation of new labels from existing vocalizations. Language Sciences 29: 1-13.

Pepperberg, I.M., Brese, K.J., & Harris, B.J. (1991). Solitary sound play during acqui­sition of English vocalizations by an African Grey parrot (Psittacus erithacus): Possible parallels with children’s mono­logue speech. Applied Psycholinguistics, 12:151‑177.

Subscribe to our newsletter