Chaser_Unlocking the Genius of the Dog Who Knows a Thousand Words

by John W. Pilley

  I was stunned.

  Rereading my paper, I confirmed that my experimental procedures in testing Chaser’s understanding of proper noun words and her ability to learn by exclusion were identical to those in the Rico study. In addition I had sharpened the paradigm for testing Chaser’s exclusion learning by establishing that she had no baseline preference for novel objects. Likewise, I had described rigorous procedures for teaching and testing the learning of common nouns.

  But I had to admit that some critical details of my studies were missing. Unfortunately I had not presented the usual tables and figures to display my findings but had only described them in words and a few key numbers. I also recognized that much of the paper was too informal and did not say enough about how Chaser attained her language learning and how I tested it. So I optimistically set about rewriting the paper.

  Although I felt my paper was comparable to that of Kaminski and her colleagues, I recognized that as a retired professor from a small liberal arts college and without a long list of peer-reviewed papers to my credit, I had nothing like their stature in the animal science community. At the time of the Rico study’s publication Kaminski led a research group at Germany’s Max Planck Institute for Evolutionary Anthropology (she has since moved to the University of Portsmouth in England), the equal of any research institute in the world in that field.

  In evaluating scientific experiments there are two major errors: Type 1 errors and Type 2 errors. Type 1 errors occur when you read too much into experimental data. That means you see an effect that has not actually occurred. Type 2 errors occur when you read too little into experimental data. That means you miss a significant effect that has actually occurred. The event is accurately represented in the data, but you don’t see it.

  As an experimenter, I was more concerned about making a Type 2 error and failing to report Chaser’s genuine achievements than I was about making a Type 1 error and overstating those achievements. But scientific journal editors and peer reviewers have to be more concerned about making Type 1 errors and publishing something that won’t stand up to scrutiny by the scientific community at large. Those two perspectives constitute one of the constant push-pull dynamics in the course of scientific progress. And science needs both. It needs experimenters to try bold things, and it needs their peers to hold them to a tough standard.

  Before long I was bogged down in my data. At seventy-nine years of age I no longer had the head for figures that I did as a younger scientist. Hearing me express my frustration about this one day, Robin said, “You should get Alliston to collaborate with you on the paper. You know he’s an expert on all that data stuff.”

  I would have loved to have his help. Alliston and I always had a great time together, whether we were conducting an experiment in the lab or going on a kayaking outing, and our families were close. Alliston and Robin became good friends when they were both Wofford psychology majors. I officiated at Alliston’s wedding to his wife, Leonor, who became one of Sally’s best friends, and Sally and I were godparents to their two daughters, Caroline and Rebecca. But Alliston had much more on his plate now than when he was my teaching assistant.

  Since graduating from Wofford, he had gone beyond me as an animal scientist in many ways. When he presented his final PhD dissertation data, John Staddon, his thesis advisor at Duke, shook his hand and said, “Welcome to the top twenty.” Staddon meant the top twenty in the quantitative analysis of animal behavior. Staddon himself was at least in the top three in the world in that field, which he’d done as much as anyone to establish.

  After teaching for several years at the National Autonomous University of Mexico in Mexico City and Eastern Oregon University, Alliston came back to Wofford to fill the spot on the psychology faculty that opened up when I retired. He had a knack for engaging his students in novel ways, such as his use of rat basketball to teach the principles of conditioning, reinforcement, and learning. He guided students in training rats to take a small ball in their forepaws, stand up on their hind legs, and drop it through a miniature basketball hoop and net. Teams of students and their rats competed in Alliston’s rat basketball tournament every year.

  Alliston was now very much a leader in the quantitative analysis of animal behavior. He was part of a changing of the guard as the researchers of Staddon’s generation retired. In fact, Alliston had just started a multiyear commitment as program chair of the Society for the Quantitative Analysis of Behavior (SQAB). Among his responsibilities was editing some thirty papers from scholars around the world for a special issue of the British journal Behavioural Processes. That work, his teaching load at Wofford, and his own research weren’t going to leave him much time to collaborate with me, and I didn’t think it was fair to ask him.

  “I’ll think about it,” I told Robin.

  Fortunately, Robin took things into her own hands. She arranged for the three of us to have lunch together early in the fall of 2007. We hadn’t been sitting down long when Robin said in her deliberate way, “I think Chaser’s learning is amazing. And, Dad and Alliston, I think the two of you could do a paper together that would have a huge impact. Dad didn’t want to ask you because he knows how busy you are, Alliston. But I believe the two of you should partner up and do that.”

  Robin tilted her face down to her plate and raised only her eyes to look first at me and then at Alliston. Following her eyes to Alliston, I was delighted to see that he was smiling happily and to hear him enthusiastically say, “I’m in.”

  “Thanks, Alliston. I’m in too,” I said.

  Unfortunately, it wasn’t until the summer of 2008 that we were ready to submit the paper, now titled “Collie Demonstrates Referential Understanding,” to the journal Animal Cognition. In the months since Science rejected the first draft of the paper, Alliston and I had improved it a great deal. I’d rewritten most of the paper and added new passages with details on Chaser’s training and testing. As I was doing that I often bounced ideas off Alliston, and he gave me many valuable suggestions for demonstrating that the tests were rigorous. Alliston also guided me in the statistical analysis of my experimental data. Finally, along with the paper itself we submitted a video of some of Chaser’s trials.

  On November 18, an editor at Animal Cognition e-mailed us that the journal was rejecting the paper based on the anonymous peer reviewers’ comments, which were attached to the e-mail.

  My heart stopped.

  And then I saw that the two peer reviewers expressed very different attitudes to the paper.

  One peer reviewer concluded, “I cannot recommend anything about this paper. The authors appear to lack an understanding of experimental controls and design in the animal language arena.” The reviewer also asserted that we should use “label” to refer to the proper noun names of Chaser’s toys rather than calling each such proper noun a “word.”

  That comment told us that the reviewer followed the school of linguistics researchers who claim that only humans can learn words and that if an animal can learn a term for something, it can’t be a “word.” I believe this reviewer was looking at the paper through such biased eyes that he or she was unable to appreciate the details of the procedures used to test Chaser’s learning.

  The other peer reviewer also saw problems with our study, and said it included too many assertions about too many kinds of language learning. Except on proper noun learning, the reviewer said, “there is hardly any systematic data.”

  However, this peer reviewer added, “[On proper noun learning] the authors provide data which is quite convincing. . . . The authors show that Chase [sic] . . . can distinguish hundreds of objects by their label. That is fascinating and a great finding in itself. However, to be able to state that Chase can also use adverbs, categories, verbs, etc., it would need exactly the same kind of systematic investigation, which I hope the authors will provide in a revised version of the MS.”

  That was more like it. This peer reviewer went on to say that we needed to do a better job of demonstratin
g that Chaser could learn a word in one trial and learn a word by exclusion, and that throughout the paper we needed to provide fuller details on Chaser’s training as well as testing. Finally, this reviewer said we had to clarify our statistical data. All this information was necessary in order to provide other researchers with the information needed to replicate our study.

  That the journal editor had given more weight to the totally negative review than to this very encouraging one angered me. It is difficult for a research finding to win scientific acceptance when it deviates from the ruling paradigm in a field. The seventeenth-century mathematician and philosopher René Descartes established one of the strongest paradigms in biology and psychology with his idea that nonhuman animals cannot reason or feel but are more or less machines made out of meat. Two hundred years after Descartes, Charles Darwin stated his belief that animals, and dogs in particular, can feel and express emotions and can reason. But anyone trying to demonstrate language learning by animals still has to battle those who faithfully follow Descartes’s ruling that animals are just meat machines.

  For a while I focused more on the negative judgments than the positive ones. I felt that we were facing a huge uphill battle to get the study published. Sally and the rest of the family grew concerned as the weeks went by and I still wasn’t back to being my usual positive self. Not even the Christmas holidays, which were always a joyful time in our family, managed to lift my mood. My play and training time with Chaser decreased. My television time increased. My social life with my Wofford peers went to zero. I skipped my morning workouts.

  My family’s efforts finally buoyed my spirits. Sally stepped up her always frequent hugs, kisses, and smiles. Debbie pumped me up with daily phone calls. Robin kept saying that Chaser’s learning was a huge deal and that Alliston and I would eventually get the paper published. And Chaser would not rest until Pop-Pop was back to normal and ready to play. Seeing that dropping toys at my feet was not enough to stop me from staring glumly at the television, Chaser put her front paws on my lap and her nose right in my face until I got up out of my chair.

  On January 25, 2009—Debbie’s birthday—I got out of bed and decided that this was the day I was going to pull myself up by the bootstraps and get back to work on the paper. The rejections had knocked the wind out of my sails, but now I was ready to accept the rejections and improve the study accordingly. If details were what the peer reviewers wanted, details they would get.

  I had to do additional tests of Chaser’s learning in blind and double-blind conditions, and record them on video. I also needed to describe my training and testing procedures more fully. Alliston had to display my data in tables and figures, along with statistical analyses of the different studies I had done with Chaser.

  Knowing that Chaser’s learning was real, I went back to the paper determined to demonstrate it with findings that gave the scientific community a high degree of confidence in her learning. The first tough decision in revising the paper was to limit its scope. I had hindered the paper by putting in my anecdotal observations of all of Chaser’s language learning, including adjectives, adverbs, prepositions, and overheard words. Alliston had earlier suggested eliminating this material, but I had wanted to be as comprehensive as possible.

  From now on the paper would deal only with the four major experiments where we could offer ample data and demonstrate Chaser’s learning in blind and double-blind conditions. The four experiments aimed to demonstrate 1) that in her first three years Chaser learned, and that she still retained, understanding of the proper noun names of 1,022 objects; 2) that she understood the separate meanings of proper noun names and commands and could correctly interpret random combinations of these two elements of syntax; 3) that she understood the three common nouns “ball,” “Frisbee,” and “toy”; and 4) that she could learn a word through inferential reasoning by exclusion. Together these four experiments matched and extended the Rico study, and satisfied the major word learning criteria proposed by Yale’s Paul Bloom and Stanford’s Ellen Markman and Maxim Abelev.

  Discussing the need for a video clearly demonstrating Chaser’s learning of more than a thousand proper noun object names, I said to Alliston, “How about we ask the Psychology Kingdom to host us at one of their meetings this semester, so we can record a demonstration with a bunch of students?”

  Alliston said, “I’ll ask Dave Pittman.”

  Dave Pittman was another student of mine who had become a psychology professor at Wofford. He was the faculty advisor for the Psychology Club, which most Wofford psychology majors participated in regularly. Alliston and I—and Dave, too, in our conversations—referred to it as the Psychology Kingdom because of a running joke that began when they were students and I was the club’s faculty advisor.

  On an outdoor trip with the Wofford Adventure Club in the early 1970s, I was jawboning around the campfire late one night with two students, Eddie Coffey and Chris Harris, who were both psychology majors. Eddie and Chris started riffing on the idea that we should rename the Psychology Club the Psychology Kingdom so people could act out their favorite personas, like being a knight errant or a wizard. I joked, “Well, if there’s gonna be a psychology kingdom, I’m the king.”

  Eddie cracked a smile and said, “Don’t agree with him, Chris, or we’ll never hear the end of it.” But then he added, “Well, Sally has thrown so many great parties for the Psychology Club that she has to be the queen of our kingdom”—he paused a beat and looked at me—“so I guess that makes you the king by default, Doc.”

  That broke us all up. By the time we saw the last embers of the campfire out, Eddie and Chris decided that they were, respectively, the prince and the grand duke of the Psychology Kingdom. When we got back to campus, they began spreading the word, and we wound up with about twenty self-styled members of the kingdom’s court. With a nod to James Fenimore Cooper’s novels and thanks to his impressive outdoor skills (studious though he was and looked, he would have been a cinch to win the Hunger Games), Alliston became the Deerslayer. Lest we take ourselves too seriously, however, whenever I was introduced as the king I always replied that I was “King of Fools.”

  Dave loved the idea of having Chaser demonstrate her learning for the current Psychology Kingdom. A few weeks later, Alliston, Chaser, and I assembled with Dave, a few other members of the psychology faculty, and about a hundred students in an amphitheater-style auditorium at Wofford. We also had all 1,022 objects in Chaser’s flock of surrogate sheep in sixteen large Rubbermaid containers.

  Dave had arranged for Wofford’s official photographer, Mark Olencki, and a local videographer to document the event. While Mark was taking his first photos of the group, Robin suggested that the students should throw Chaser’s toys in the air so he could get a dramatic action shot. Chaser enjoyed the spectacle, and she was in her element meeting the students and getting attention from them.

  A chaotic scene became even more chaotic as ten students in teams of two rummaged through the 1,022 toys piled on the auditorium floor. The five teams each picked ten toys. The selection took a few minutes because the students weren’t content to grab just any toys. They competed to find the most interesting and cutest toys.

  The selection of objects was random. In addition, none of the students had ever worked with Chaser and none of them knew the names of the objects prior to selecting them. The teams made lists of the names written on their ten objects and gave the lists to me. Each team in turn then randomly placed its chosen objects on the floor behind me so that I could not see the objects when I asked Chaser to retrieve them by name. These procedures constituted a double-blind test of Chaser’s learning of the toys’ proper noun names.

  The fifty objects in total included some whose names Chaser had learned as a young puppy, some she’d recently learned, and some she’d learned at different points in between. In at least seven cases I was not sure what object a name referred to, because I could never keep all of the 1,022 objects and their names straight in my own
head.

  Chaser’s memory for them was better than that. Following her retrieval of the objects, the five teams of students evaluated her accuracy. She had correctly retrieved forty-six of the fifty objects, or 92 percent. The entire Psychology Kingdom cheered Chaser’s success.

  Dave Pittman opened up the meeting for questions, and John Lefebvre, newly appointed chairman of the Psychology Department, asked if Chaser would retrieve objects for people other than me. I invited John to come down onto the stage and find out for himself.

  John is not bashful, and in a minute he was on stage. Having sat in on several of John’s classes, I knew he would test Chaser’s limits. I threw eight of Chaser’s toys on the floor, and asked John to have at it. He picked the smallest stuffed animal, a dog named Tiny. And then, out of Chaser’s sight, he threw Tiny several steps up in the amphitheater. He turned to Chaser and said, “Chaser, find Tiny.”

  Chaser quickly nosed through the objects on the floor. No Tiny. Although I thought the test was a little unfair because Chaser was in a strange environment and did not know John, I said nothing.

  John repeated, “Chaser, find Tiny,” and Chaser began to explore the room. Several times she reexamined the objects on the floor. John wisely continued to repeat, “Chaser, find Tiny.” At least two minutes passed while Chaser looked all over the stage. Finally she approached the steps that led up through the rows of seats in the amphitheater, saw Tiny, and picked him up in her mouth.

  Realizing that I was holding my breath, I heaved a huge sigh of relief. Again John showed his knowledge of dogs by enthusiastically praising Chaser. And the Psychology Kingdom of students clapped and roared their praise as if they were at a Wofford football game.

  Dave Pittman was looking at the clock, but before he could bring the meeting to a close, a student asked for one more demonstration of Chaser’s word learning. I said, “Chaser has learned the meaning of ‘clean up,’ so I’m going to ask her to do that now. You will see, however, that she is like a two-year-old child and will need reminding to finish the task.” I turned to Chaser and said, “Chaser, clean up.” She quickly put two toys in the plastic tub but then dropped one at my feet. “No, Chaser,” I said. “It’s time to clean up.” With that reminder, and a few more before we were done, Chaser put all eight toys in the tub. As the students clapped, a girl shouted, “Can Chaser clean my room?”