Bibliofreakblog

Interview with Dr. Irene Pepperberg

It’s tough to say who is more famous–Dr. Irene Pepperberg, or the late Alex the grey parrot.

I’ve read about Alex in many linguistics books, and leapt at the opportunity to interview Dr. Pepperberg, who worked with him for 30 years.  

Of course, like all interviews I have ever done, I thought of other questions for Dr. Pepperberg after I already sent my questions to her.  Damn.  

But since I sent her fifteen questions, it’s probably good that I didn’t include any of the ones I thought of later.  Without further ado:

1.You came from the “hard sciences”–specifically chemistry.  Do you think that this has helped or hindered you at all?  I could see that you might be able to break out of the box easier (certainly you were never constrained by trends in Behaviorism) but at the same time, not having a background in linguistics or animal behavior might have had its setbacks.

Having a hard science background did work both ways. On the positive side, it meant that I had a very strong grounding in experimental design, in setting up controlled studies, and in finding any weaknesses in a project. And, as you mention, I hadn’t learned that what I wished to accomplish was supposedly impossible. On the negative side, I did have to train myself in psychology and linguistics, neurobiology and ethology—which meant enormous reading lists for a very long time. Too, getting a tenure-track job in either psychology or biology with a degree in theoretical chemistry (even from Harvard) was a challenge.

2. When Alex first learns to put more than one adjective (or category) with an object, he calls a green clothes pin “green wood peg wood” (peg wood being his word for clothes pin).  This was an amazing leap for Alex.  You write that the “perfect response would have been ‘green peg wood.’”  Why is that?  How does repeating the noun detract from his achievement?

Had he said “green peg wood”, he would have shown that he understood how to segment “green wood” into “green” and “wood” and thus that “green” was an independent entity, referring to color, separate from the material. What he did was show that he knew something about “green” with respect to referring to color, but not (at that time) that it was specifically modifying “wood”. His behavior, however, was quite important, because it showed that he understood that he should combine labels to refer to a novel entity.

3. Alex coined many of his own terms and phrases over the years, such as “cork nut” for almond, because he thought its look and texture was a bit cork-like, and “banerry” for apple, as it looked like a cherry and tasted to him like a banana.  You mention that this had previously been shown in apes.  Washoe the chimp came up with “water bird” for swan and Koko the Gorilla created “white tiger” for zebra.  When my niece was about 18 months old, she came up with “light bee” for firefly, which amazed all of the grown ups around her.  But I was thinking, is this just another form of segmentation, combining two categories in a new way, such as “green peg wood.”?

In some ways, you are correct…but in other ways, not quite….What the child and the nonhumans did were more sophisticated…In the case of “light bee”, your niece took two very different concepts/labels—her knowledge that a flying insect of roughly that size was usually called “bee”—and the action of it emitting flashes—to form the new label for a totally novel entity. She probably knew it wasn’t really a bee, but took the closest possible label. Ditto for Alex’s actions with “banerry”. In contrast, combining “green” and “peg wood” is simply using a modifier to add information to an existing, known, labeled entity.

4. You spent some time at MIT with Alex as well as the other parrots.  MIT is the linguistic haven for Noam Chomsky and those in his camp.  Did this ever have any influence (for good or bad) on your work?

When I was at MIT, I was in the Media Lab, which is a somewhat separate entity…although I had some limited connections with the Brain and Cognitive Sciences Department and the AI (Artificial Intelligence) group, during my time at the Lab I was primarily working on devices such as “InterPet Explorer” (a limited web-browser for parrots) to enrich the lives of companion and zoo animals.

 5. When Alex first learned shapes, he learned them by number of corners, so that, for example, a square was “four-corner.”  Why did you choose to teach him this way?  Do you think it helped when he started learning numbers?

Well, when you think about our English labels….”triangle” is really just a form of “three-angle” or “three corner”; “pentagon”, “hexagon”, etc. also refer to the numbers of angles (from the Greek, gonia) or corners of the shape…“square” is actually an anomaly…so I decided just to use English ‘translations’; I chose “corner” rather than “angle” because I preferred the way it sounded..…

6.  A basic understanding of same vs. different, blue vs. green, bigger vs. smaller, more vs. less, is evolutionarily helpful in the wild, even if parrots don’t have “labels” for such distinction.  At the same time however, by giving these categories labels, you impose a certain abstraction on them. How much do you think that you trained Alex and how much do you think you brought out of him?  

As you note, I don’t think we could have trained Alex to label concepts unless he already had a basic understanding of these concepts—what I like to refer to as an underlying cognitive architecture. But we know from studies on humans (children and even primitive tribes) that providing specific labels provides the ability to think about these concepts in a far more abstract manner. A good example involves number concepts—we know that any animal that forages and that very young children have some concept of “more” versus “less”—and we know that hunter-gatherer tribes like the Pirahã  have labels that refer, roughly, to “very, very, very few”, “very few”, “some” and “many” because that’s all they need. But clearly a Pirahã child who was placed in Head Start would learn number concepts just as any other child so exposed…thus labels for exact quantities allow for an understanding of exact quantities, showing how the combination of underlying cognitive abilities interacts with language to promote more abstract cognition—in humans and nonhumans.

7. You didn’t teach Alex much grammar, and yet sometimes he would break out in a full sentence, such as when he was frustrated with you and asked, “Well, what DO you want?”  Do you think he understood concepts of grammar, such as verb tense, articles, and prepositions?

No, I don’t think so. When he used these complete sentences, he was generally using them in contextually applicable ways—he understood the general sense of the sentence and the appropriate context for use, but not necessarily the meaning of each of the individual words. In contrast, he did understand simple sentence frames… “I want X”, “Wanna go Y”…and these were used intentionally…if he asked for X and you gave him Z, he tossed Z back at you. We were working with “over” and “under” when he died.

8.  It seems to me that Alex demonstrated not that animal minds are human-like, but that human minds are animal-like because he demonstrated a number of abilities that are sort of like the building blocks of language.  Because humans are the only living species we know to have developed what we consider language, it seems almost like these building blocks were just waiting around for us to make use of them–kind of like how we have brains that can learn to read or use computers, because of what different traits were already there.  How do you think these building blocks were/are advantageous to non-human animals, who are not using them towards language?  What I mean is, where is, for example, Alex’s ability to form “anticipatory co-articulation” (that is, pronouncing the “k” sound in “corn” differently than in “key” because of the sounds that make up the rest of the word) beneficial?

I believe that some of these traits are likely useful in their own communication systems. And although I agree that no nonhuman has acquired human language, I also caution us not to dismiss the complexity of various animal communication systems. We have not yet entirely deciphered such systems. Think about birdsong….after about 50 years, for some species we now understand a bit about which songs are used in territorial defense and which ones are used for mate attraction, and something about how different songs may express different levels of aggression… but if you slow most birdsong down (e.g., record at 78 rpm and play back at 33 rpm), you hear an incredible complexity that you don’t hear normally, and I have to (a) assume that the birds’ ears and brains process this complexity and (b) wonder  why that complexity exists if it doesn’t have meaning for the birds—some meaning we don’t yet understand.

9.  Whatever happened to Kyo?  You mentioned that he had what seemed like an avian version of ADHD.  Did you leave him in Arizona because he was just too hard to work with?

When I moved East, and knew that all three birds would have to share one room, I realized that Kyarro would not have the separate space away from the two other (somewhat aggressive) males that he needed. He is living with a friend of mine in Arizona who has two female Greys, with whom he seems very comfortable.

10.  Alex and Griffin’s (another African Grey) developed a game while you shared food with them, where Alex would say “green” and Griffin would reply “bean”, repeatedly.  Of course, they seemed to just be playing, but is this sort of an avian version of what anthropologists term call and response?

In the wild, we know that mated pairs of parrots share duets—particular sets of vocalizations, unique to that pair, that they use to identify one another across the savanne and/or when in large flocks. Quite possibly, Griffin and Alex adapted this behavior into a “buddy” system in the laboratory.

11.  Alex only ever called one human by name.  Why is that?

I believe there were several reasons. First, we often had several students at a given time with the same names…we might have two students named Denise, or two named Steven/Stephen; one year we had several “Chris/Cris/Kris/” students, male and female…so Alex quickly saw that the names were not individual identifiers, and I’m guessing he figured they would not be useful to learn. Second, although many students stayed for several years, some did not…so there was little reason for Alex to learn their names. And as for me…students varied in how they addressed me…calling me “Irene”, or “Dr. P”, or “Professor”, or variations…so likely Alex didn’t view me as having a single identifying name, either. Alex did take a particular liking to a few students, but the only one with a truly unique name was Spencer, whom he called “sir” (likely the last bit of the name).

12. Griffin picked up Alex’s pronunciation.  Is this because he was still more inclined to speak like a member of his own species than humans, or because it was just easier to mimic those sounds?

We don’t really know.

13.  Was there any sort of fall-out after Alex’s death for the other birds in the lab?  Did they demonstrate any sort of mourning or refusal to work?

It has been difficult to separate Griffin’s and Arthur’s own responses to Alex’s absence from their responses to OUR behavior after Alex’s death. Clearly, the students, my lab manager, and I were deeply affected, and responded by acting in ways that, in the view of the two remaining birds, were not normal. I do have to say that, probably because Alex interrupted all of Griffin’s and Arthur’s sessions, often giving the answers first, Griffin still sometimes looks towards Alex’s cage for the correct response before replying himself.

14.  Where is your research taking you now?  You write that the next step with Alex may have been to work on optical illusions to see if animals “see” the world as we do.  Will you be able to continue that with Griffin or Wart?

 That is what we are planning on doing. But first they have to learn a bit more.

15. I’ve read about your research with Alex in many books on linguistics.  What books do you recommend for people who want to learn more after reading Alex & Me?

To understand the science, people should read The Alex Studies. That book describes the first twenty years of research, the rationale for the research, the training methods… I’m not sure if I’ll write “volume two”, to cover the next ten years, but we’ll see. Of course, the original research papers are available from university libraries and interlibrary loan if anyone wants to read them.

I am so grateful to Dr. Pepperberg for taking the time to answer my questions!