 William Rumbau is Director of the Language Research Center at Georgia State University in Atlanta, where he serves as Regents Professor of Psychology and Biology. He earned a Bachelor of Arts at the University of Dubuque, a Master of Arts from Kent State University, and a PhD in General Experimental Psychology from the University of Colorado. He was the inaugural G. Stanley Hall Lecturer on Comparative Psychology for the American Psychological Association in 1984, and was President of the APA Division on Comparative and Physiological Psychology in 1988. Last year he served as President of the Southern Society on Philosophy and Psychology. Over the course of 38 years, first at San Diego State, then at Emory University's Yerkes Research Center, and for the past 25 years at Georgia State University, Dwayne Rumbau has researched the relationship between primate brain and behavior, especially the ability of primates to learn complex tasks, in other words, intelligence. Today, Dwayne Rumbau will present a case for the need to alter our perspectives on non-human primates and apes in particular. His talk will reflect his recent research, focusing in part on shifts in the learning process, which have prompted him to ask questions such as, why do apes learn better than monkeys? And why do some humans learn better than apes? I spoke to Dr. Rumbau on the telephone recently to see if there was anything in particular that he wanted me to be sure to include in this introduction. Without hesitation, he humbly said three words, keep it short, so I'll stop here. Speaking on the psychology and intelligence of human, ape, and monkey, please welcome from the Language Research Center at Georgia State University Dr. Dwayne Rumbau. Thank you very much for that very generous introduction, and thank you Richard Elvay, thank you President Stroyer, and thank you Gustavus for exercising leadership in the formation of these Nobel conferences. My topic today is on human competence and primate roots and really the foundation of intelligence. Humans really have not been comfortable with the idea that they are part of nature. I'll take the first slide on here, or that we have some kind of mental relationship with animals. Historically we have searched for characteristics that definitively would separate ourselves, even from those species, those forms most closely related to us, the primates. But so much has been learned in the past 30 years that we now need to redefine our relationship with them. Of the over 200 species of primates, those most similar to us are the great apes. The orangutan from Asia, and from Africa, the gorilla, the chimpanzee, and the bonobo. And now we have them all in one slide, an unusual picture that shows these as infants, and this is the way most people think of the great apes, but actually they grow to be very large and strong creatures. Of the apes, the ones most familiar to us are of the genus Pan, the so-called common chimpanzee, and the not so familiar bonobo. Those two in the middle, the gorillas on the left, bonobo, is next to the right, common chimpanzee and orangutan. But before discussing work with them, I want to tell you something not so much about a gorilla, as about how one was to conduct research 35 years ago. I had set about to study the complex learning skills and intelligence of apes and monkeys of the San Diego Zoo. One of my most favorite subjects, at least potentially so, was a silverback male gorilla named Albert. I had designed and built an apparatus that was essentially a Wisconsin general testing apparatus. And this apparatus of Harlow's was made of wood and wire, and would not do for my purpose, because my subjects included adult apes of extraordinary strength. I chose a stouter material, quarter inch thick steel plate. The bins that held the objects for discrimination learning research were of stainless steel and mounted so that the food for correct choices could be obtained by their displacement. The fronts of the bins consisted of bulletproof glass, 3 eighths of an inch thick. The apparatus was strong, but as it turned out, not strong enough. Albert was 17 years old at the time, and being a gorilla, he had never gone to the public school. As it turned out, he didn't even like school, not even one bit. He had what we would call a bad attitude, so to speak. Every day among test sessions, he would display dozens of times in the classic pattern of wild gorillas, with chest beating, screaming, charging. His displays ended by crashing into his side of the test apparatus. He did this with such strength that paint literally peeled off the wall and drifted down into my hair. Frequently, I feared that his side of the apparatus was about to become my side. Just listen to him. Here he comes. He gave those kind of displays about 30 times every hour. Over several months, however, he settled down and I began his first training, which consisted simply of having him push back on whichever glass-fronted bin held an object. But then the day came on which he had to make a choice between a pair of objects. Each object in its own bin. For the first trial, he was correct and got some nice grapes from the food well. On the second trial, however, he was incorrect and got nothing. That this disturbed him, as I knew that it did, became apparent on his third trial when, once again, he was incorrect. For Albert, two errors proved to be just too much. Albert was now frustrated, if not out and out, angry. As I tried to raise the safety door to prepare for the next trial, he forcibly re-opened it with a single finger. His face scowled and his brow furrowed. Beads of perspiration broke out all over his face and the distinctive pungent odor of upset gorilla permeated the air. He quickly made a fist with his right hand and then delivered a blow to the glass-fronted bin on his right. Just look at the damage that it did. The silicone on his floor left an impression of his hand, his fist, even his thumbnail when they hit on his wrist. Albert had discovered that the apparatus had a glass jaw and one by a knockout. He had broken it. Better it's jaw than mine, however, for it too is of glass. Albert was unscathed, but never did graduate. I now continue with quite a different and more recent story about pansy. And pansy is a chimpanzee. The story is about what I heard her say. If we are to learn from chimpanzees by listening to them, how can that be done? After all, chimpanzees, though noisy, can't talk as we do. So how can we listen to them? The answer, though not simple, is that not all communication is by speech and hearing. We also hear a great deal with our eyes as we watch behavior. The story entails a jeep. I had just bought an old jeep in 1977, in part because it was in such good shape that its black-padded dash was perfect. Not a blemish on it. Sue and I were in the jeep with pansy and panbenesha showing here, with pansy punching on panbenesha's tummy having a good time. When they were about two and a half years of age, driving over the rough terrain that surrounds our center, we had stopped the jeep and walked about the hills and woods and were preparing to drive on. As we settled into our seats, my eye caught something new, a scar on my jeep's black-padded dash. They looked like teeth marks. Closer inspection confirmed it. They were teeth marks. Sue, how did these get here? Her reply, pansy did it as she came across your seat. You have to watch her. My eyes focused sharply on pansy as I asked, pansy, did you do this? I'm really disappointed in you, all the time pointing to the bite on my black-padded dash. I really didn't expect to be heard and I certainly did not expect an answer, but I got one. Throughout the lecture, she looked me squarely in the face and didn't blink once. Only when I was done did she answer. She began by taking my right hand with her left hand. Turned my palm up. Then, without breaking eye contact, she brought her right hand to mine and pressed something into it. She then gently closed my fingers around it and then firmly thrust my hand to my chest, all the while sustaining eye contact. I then opened my hand and found a beautiful wild flower, one that she had picked, I guess, while we were walking. What was she saying? I believed that in her own way that she was saying, here's a flower that I picked. I want you to have it. I'm sorry. Don't be angry anymore. Whereupon she wheeled and looked out the windshield as though to say, life goes on, let's go. I was so impressed that her teeth marks on the dash no longer mattered. Be clear pansy had not been taught to make up by doing what she did. Her behavior, her answer, was an invention of the moment, her invention. I hope that you agree that her answer was more eloquent than spoken words ever could have been. I also hope that the point intended by my account of these two incidents is obvious. Thirty-five years ago, I placed bullet-proof glass and steel between myself and Albert. I hadn't taken time to make a good friend of Albert and nothing went well with him. His emotions denied him the opportunity to excel, as I'm certain that he would have. By contrast, pansy had become a friend, a companion. The only steel about us was in the jeep. Time has served to change our view of apes. As stated earlier, we are now at a point in history when we need to revise a long-standing perspective of ourselves in relation to other primates, and notably the great apes. A revision is needed because of the substantial psychological as well as biological continuity between us. The degree of continuity is a function of genetic relatedness. Hence, the psychological continuity is more pronounced between the great apes and us than between any of the monkeys and us. And significantly, the continuity is closer between chimpanzee and us than between chimpanzee and gorilla. That's in terms of DNA. Primatologists have learned that many of the hell-to-be human qualities are not, in fact, unique to us. Many of them are general states and processes shared by other primates, if not by animals in general. We now know that apes are capable of symbolic thought, basic dimensions of language, elemental numeric skills, and planning. And while the great apes are not our ancestors, they and we share common roots. So it follows either that the apes are not the unknowing and thinking and feeling beast machines that Descartes declared several centuries back that they were, or that both they and we are. Frankly, I don't believe that either they or we are beast machines, and that it is timely to revisit even the basic terms of behaviorism that have discounted the roles of sentience, rationality, and feelings, particularly in animals. Let us continue by considering language. Traditionally, language has been viewed as speech, a unique endowment of our species. When did your child begin to talk? Is a common question of parents. Chomsky holds that language is unique to humans and that its foundation is some genetic mutation. More recently, language has come to be viewed as a complex system of learned behaviors. This view holds that language is the culmination of processes that have their roots in the evolution of the brain and the emergence of intelligence, and that the foundation of language is comprehension, not speech. From this perspective, primates are attractive subjects for research, and that they have been for several decades, and now we shall see that more clearly. With regard to apes and animals and language, let us observe that the last third of this century, the 20th century, has seen a great increase in language research with animals. Good reports include dolphins carrying out novel requests, sea lions learning the meanings of new manual signs standing for words apparently on the basis of inference, a parrot seemingly able to answer questions regarding multiple characteristics of objects, and great apes learning various kinds of symbol sets. A special focus here is the capacity of bonobos to understand human speech, not only single words but sentences as well. Notably, Kanzi can understand speech at a level that compares very competitively with that of a two-and-a-half-year-old child, and that's a very conservative estimate, it's probably more like three-and-a-half perhaps higher. The bonobos' special capacity to learn language observationally was discovered quite fortuitously in efforts to work with language instruction with Kanzi's adoptive mother, and this is Kanzi's adoptive mother on the right. She failed to learn language by formal instruction, probably because she was quite mature when captured and brought into captivity by another institution initially for studies of reproductive biology. Though Matane did not benefit from training, her adopted son, Kanzi, learned it all. Just by being a playful observer, he learned what she did not, that he had learned what he had did not become a parent until he was two-and-a-half years of age and his mother was separated for purposes of breeding. More than 400 novel sentences of request in controlled tests were given to Kanzi at eight years of age by Dr. Sue Savage Rumba and her colleagues and to a child, Alia, two-and-a-half years of age at the time. Alia's mother worked with the apes of our center. Let's just take two examples of sentence comprehension briefly. Here, of the set of more than 400, Kanzi was asked, make the doggy bite the snake. From the upper left, you see a sequence running across the rows to where Kanzi picks up the dog and then the snake. He opens the dog's mouth, puts the snake's head into the dog's mouth, and thrusts it down further and further with his thumb. He was not trained to do this. He had not seen it done, but he did know what a snake was and what a dog was and what bite meant. To make clear that this was not an easy sentence, let's see what Alia did. When Alia was asked, make the doggy bite the snake, she first of all appeared to be puzzled, upper right-hand frame, finally went over to the doggy, leaned down gingerly and bit it herself. Kanzi was 74 percent and Alia was 66 percent on these questions in that they answered promptly without further restatement. Sue Savie-Drumbaugh will present more on Kanzi's sentence comprehension tomorrow including exciting video. Today we will see only his single word comprehension as portrayed by video. Turn up the sound, please. Does he really understand what he hears? Let's watch more. Kanzi, see if you can find mushrooms. Mushrooms, mushrooms. Right, go to the mushrooms, real good. Can you turn my camera? Okay, now, now. Okay, you're doing real good, Kanzi. See if you can find maju, the orangutan. You see maju? Good job, good job. See if you can find some melon, melon, melon. Thank you. See if you can find green beans, green beans. Very nice. See if you can find a picture of Sue. Sue, very nice. Thank you, that's me. See if you can find a picture of coconut, coconut. Good, good job. See if you can find a picture of oranges, oranges. You see a picture of oranges? Thank you very much, Kanzi. See if you can find a picture of banana, banana. Very nice. And now we need a Paminatia. Can you see Paminatia? Another bonobo. And now we need keys. Do you see the picture of keys? Thank you, Kanzi. Obviously, Kanzi can choose correct pictures in response to Sue's voice. Be clear that Kanzi was not trained to do this on a trial-by-trial basis. He hadn't even seen those pictures for several years. They were randomly selected by another person and randomly placed on the board. This is the first time he was also tested with that particular board. Kanzi did not acquire his language and speech comprehension through standard operant conditioning procedures, that is, by executing motor responses and receiving pellets or other reinforcement for them. Rather, he learned by living, by observation. He did not learn how to make specific responses to specific stimuli. Rather, he learned of the predictive relationships between all kinds of things and events of his world and learned words for them. It was the design, the structure, the function of his fine primate brain that brought organization and sense-making to his perceptions and experiences. Significantly, the information Kanzi thus accrued had remained silent for several years. But in due course, it became active and useful. We leave Kanzi for now by noting that he also has learned, by observation, how to nap flint, to make tools to get things from puzzle boxes secured by rope or leather. Kanzi is quite sophisticated and now even turns the cobble so that it presents its best edge for producing a large sharp chip when struck. Each chip is assessed for its sharpness and appropriate size. And the thicker the cable of rope to be cut, the larger the chip he produces, letting the smaller ones fall by the way. Once he gets that rope cut, he will have access to a prized incentive in that box. Because the brain is vital to all of these magnificent behaviors, let's briefly consider it from an evolutionary functional perspective. Large and complex brains are of high metabolic cost. For them to evolve, they must have made a real difference in survival and reproduction. Large complex brains, presumably, were selected for their enhanced learning and memory, which make it possible to carry forward important lessons of the past. The capacities transfer even small amounts of learning to a leveraged advantage in an ever-changing environment also was surely important. Compared to apes, monkeys have small bodies and even smaller brains. In other words, if monkeys were just small apes, they would have larger brains than they do. Although they are not as skilled in learning and the transfer of that learning to new situations, they can do some really smart things. Let's consider some of them. The WGTA discussed earlier is now replaced by our Language Research Center's computerized test system. Studies with it have substantially upgraded our estimates of Rhesus monkey intelligence. Here you see the Rhesus reaching through a portal, getting food after having used the joystick which is placed horizontally immediately below the monitor. We have learned all kinds of things that Rhesus can do, none of which the WGTA literature would suggest. Rhesus can use a joystick-controlled cursor to make their choices in learning tasks. They can learn to differentiate between two-dimensional patterns presented on a monitor and as presented now on the screen, they can chase targets adroitly and track them and even lead them as they shoot. They shoot them down with bullets of light that are aimed by the directional throw of the joystick by their hand. They make allowances for the trajectories of the targets and the speed of the targets, even for the velocity of the bullets they shoot. Thus the Rhesus makes skilled predictions in a complex computer-controlled system, something only humans were thought capable of doing up to about six or seven years ago. Let's watch a Rhesus monkey work briefly in our system. You will see the screen from the monkey's view and how the monkey worked on it. Video please. See the cursor moving on the screen. The monkey is doing that through use of the joystick. Those patterns are very difficult to discriminate between and yet the monkeys can do it, whereas in the old WGTA they had a terrible time doing that. There's the joystick and here's the pursuit task. Monkey has to capture that circle and then stay within it for a predetermined number of seconds before it falls out the clock resets. They can do that for 45-50 seconds. Video please. Video. So they are able to do really swift things that never could have been assessed in the WGTA. An important point, technology does enhance our opportunity to ask better questions. Rhesus, to our surprise, also really learned the relative values of the numerals zero to nine. And when given array of random numbers, in final tests they very reliably would count down from the highest to the lowest for corresponding quantities of pellets. In other words, here, given an array of five numbers randomly selected zero through nine after they had learned the relative values of the numerals by picking one or the other of each pair in turn and getting a corresponding number of pellets associated therewith, they, in this test from trial one on, started with the highest one, eight, and then would go to six, and then to five, and then four, and then to three, and so on. They didn't have to learn this kind of complexity, but they did. They could have selected these numerals in any sequence and obtained the same total quantity of pellets as they went along. But yes, they did count down nine, seven, nine, eight, seven, six, and so on as though they were counting down for a lift off at a NASA launch site. Now you know what NASA is doing as they count down before the launch. They were super learners. They learned a lot more than what they had to. They learned a lot of things we never thought they could do. Rhesus also worked better if they can choose the task on which they work. Even for them, control of their world is important to their motivation and quality of work. Our test system serves an important object lesson. Science can be no better than methodology, and our system has given us a new and expanded perspective of brain and intelligence, even with monkeys. Rade Apes' brains are very large relative to those of monkeys, even when body sizes are taken into account. This is even truer for the human brain, which is three times larger than the apes, even though our bodies are not that different in size from theirs. And the basic architecture of the brain remains basically the same from monkey to ape to human. That's not to say that there aren't differences in the levels of complexity and how the associative networks are put together. Large brains enhance the ability to learn complexly, to learn of relationships between all kinds of things, and even to transfer small amounts of knowledge to extraordinary gains. From the slide now on the screen, I want to make a single point. The larger the primate species brain is, the more likely it will be able to use high levels of learning to an advantage in transfer tests. The higher of these two curves is for vervet monkeys, not too unlike the rhesus monkey that you saw working on our test system. The brains of the vervets are relatively large and complex, especially when compared to lemurs. The lower curve, the one that drops off to the right, is for lemurs. As one moves to higher and higher levels of learning, that is to the right end of the baseline, one sees the net effect of that learning on transfer tests. Increased task mastery, that is learning. Enhanced transfer for the vervets, but not for the smaller brain, more primitive primate form, the lemur, that is not even a true monkey form, though it is a true primate. This slide portrays the same principle across 12 primate species, ranging from prosimians on the left to monkeys and to great apes. Only the larger monkey species and the great apes could use increased amounts of learning to a leveraged advantage in transfer tests. And it is for this reason that I believe Curlers chimpanzees were able to be so clever. But before we look at Curlers, let me point out here that the shift from left to right tends to be a gradual one. And that gradual enhancement of transfer skills is associated with body size, brain size, the amount of absolute brain that the primates get in addition to the fact that they are mammals and primates. As we said earlier, the great apes benefit more in terms of brain evolution than have the rhesus. And so when Curlers started his work at the turn of the century with chimpanzees, it was no coincidence that they were able to do remarkable things to stack boxes used poles to get bananas otherwise out of reach. Everyone recognizes marked similarities between the appearances of apes and humans. Accordingly, it is reasonable to expect that apes and humans also might share impressive psychological competencies. This surely was the basis for use of apes earlier this century. Let us consider that history provided the foundation for present-day science. Constraints of speech by apes became apparent as the haze with their chimpanzee Vicki tried to teach her to talk. They did use operant conditioning procedures by requiring a little bit more of her as she tried to utter each word. And by teaching her specifically how to use consonants, they finally got her to give whispered approximations of four words. Mama, Papa, Cup, and Up. In this very rare film that we will now show a segment of, you'll see Vicki's best efforts to talk. We'll have the video now, please. Vicki has learned by imitation to incorporate two of these sounds into whispered approximations of the words Papa and Cup. Now, at the age of two and one-half years, she is learning to use all three of her words appropriately in solicitation. And wait, wait. Now who am I? Papa? Just trying to say Papa. Yeah. Notice how she uses her hands. You see, it was a very belabored action on Vicki's part, and not infrequently she went through her entire list of learned utterances trying to satisfy the one appropriate to the trial. So to say that she talked is really being extremely generous. Fundamentally, she learned some very whispered approximations of words, never came to talk in a way that could allow her to converse. Vicki's use of hands as she attempted to talk encouraged the gardeners of Nevada to use American Sign Language with their chimpanzee Washoe. Washoe's contributions were several, among which were her mastery of a large number of manual signs, some of her own innovation in generalized use, and her talking to herself as she looked at magazines and played. Through these and other contributions, Project Washoe revitalized interest in Ape's potential for language. A second effort of the mid-1960s was by Primac, who used various plastic tokens to function as words for his chimpanzee, Sarah. Project Sarah's contributions included the suggestions that chimpanzees might attribute states of knowledge to others. In 1970, my colleagues and I designed the Lana Project. Unique to this effort was a computer-monitored keyboard. Each key had a distinctive geometric pattern called a lexagram. Those symbols in the row of projectors are lexagrams. Each lexagram was to serve as a word for Lana, a chimpanzee. The main goal of the project was to determine whether a computer-controlled system might be perfected to advanced research where language abilities were limited, either due to relatively small brains reflecting genetics, as with apes and monkeys, or brain damage, as with children with mental retardation. Additionally, we set about to objectify training and data collection. Initial training of Lana taught her a variety of stock sentences that would cause the computer to activate several devices that would give her sips of various drinks, foods upon request, music, a movie, or even a slide. And here she's looking at her prized M&M. She'll notice she's pointing at them. She can recognize what they are. And she looks at her ball, which she also enjoyed playing with. Or she could ask for human company and grooming. Lana liked and still likes coffee. And when she learned to ask us for coffee in a variety of ways, such as, you give Lana a cup of coffee. You give this that's in cup to Lana in cage. Really impressed us. Let's watch Lana work while she was about four years of age. And you'll see her performing two stock sentences, the first for a piece of apple and the second for juice. There's no sound on this. Let's have the video, please. Okay, here's Lana on her bench. She had to pull down on the overhead bar in order to turn on the keyboard. That was to keep her from trying to use two hands at one time. You see, she was very skilled in finding and selecting the keys. And she got a piece of apple. Please machine give piece of apple is what she is using on the keyboard. And now she's asked, again, please machine give piece of apple. And you see how it looked at her on her projectors. Now she's getting juice, having asked, please machine give juice. Thank you. Lana was very good at using dozens of stock sentences and even extended their use to novel problems that she faced. When a vending system for food jam, for example, she came on her own initiative to ask people to move behind the room. By name, for example, Bev moved behind room. Then promptly used still another stock sentence that would normally have activated the vending device that was malfunctioning, such as please machine give piece of banana. Then by pointing her finger at it, she seemingly said, see, it just isn't working, whereupon then the technician would fix the problem. Although she used her lexicrams in complex strings, her comprehension was limited to stock questions, such as what name of this that's red. And Lana would look at six objects, discern the one that's red, and then answer, for example, shoe name of this that's red, or bowl name of that that's red, depending upon which object was, or what color of this shoe, whereupon she'd have to discern the shoe from the bowl and ball and the cup and so on, and then determine what color it was and say a color of this shoe red or color of this shoe blue or purple, orange, white, or black. Lana creatively asked for various items for which she had not been taught the names. For example, she asked as a cucumber, as the banana, which is green. For an orange-colored Fanta soft drink, as the Coke, which is orange. For an overly wrought banana, as the banana, which is black, and so on. And even though she did not know that orange was the name as well as the color of the citrus fruit orange, she solved the problem by using what she knew to be the word for apple. When asked to name the fruit, ask for it. She asked for it as the apple, which is orange. All of those are extensions well beyond anything that she was taught specifically. Lana is now 27. She's been a very good mother and works frequently on her counting tasks. In response to seeing either a one, two, three, or four on a monitor, as you now see on the screen, she can remove the appropriate number of boxes from the array of boxes at the bottom, random in quantity, random in arrangement, each trial, with the target numbers one, two, three, four, randomly sequenced. To do this, she has only her memory to guide her, for there is no cumulative record of her counting within the course of trials. Somehow she must monitor how many have I removed, how many in addition need to be removed before the trial can be completed correctly. With about 80% accuracy, Lana does this. She counts. And when during the course of a trial, she changes her mind, and instead of ending the trial, goes back and picks up another box, or instead of picking up another box, goes and ends the trial, she significantly increases the probability that she will be correct. The moral of this is when Lana changes her mind, it's probably for the best. The Lana Keyboard served as a way of extending research and assistance to the benefit of special populations. Thus, in 1975, we launched a study at the Georgia Retardation Center in Chambley, Georgia, and worked with children and young adults who otherwise had not learned any language from other previous course approaches. They learned upwards of 70 or 80 symbols, and one chap even learned to carry on simple, meaningful conversations. Still, another major project of the early 19... I should say this is a commercially made board now that's about the size of a book, weighs about three pounds, and is in frequent use by children across the country. Still, another project of the early 1970s was terrorist research with NIM at chimpanzee. The project started with the high hope that NIM might even report on certain aspects of his private life and views. Wouldn't that be wonderful? The project ended, however, when terrorists concluded that the majority of NIM's signing was simple imitation of what he had seen his mentors sign recently. NIM was concluded to have no language and a cloud formed over the field. But not all analyses supported terrorists' conclusion that apes were only senselessly imitating their teachers. For example, Miles' Chantek. His signs were not attributable to imitation in any large amount, and were appreciably more spontaneous as well than were NIM's. Similarly, Lana only infrequently, that is less than 10% of the time, used lexograms that had been used by her experimenters within the 10-minute timeframe that preceded her own novel utterances. Quite independently, the role of imitation in language acquisition was addressed by researchers of child language acquisition. They concluded that imitation was indeed a normal and effective means whereby children both learn and affirm the appropriate use of words. Thus, from this perspective, to the degree that NIM did imitate in terrorists' laboratory, it might have served to instate skills that became accessible and accessible and useful to him once he had been returned to the place of his birth in Oklahoma. Thereby published accounts, he was a competent, effective signer. At the same time that project NIM was ending, an important study with Sherman and Austin was reported by Savvy Drum Ball. This study concluded that Sherman and Austin had mastered basic semantics. Semantics, that is, word meaning, is the most basic building block of language. Unless words have meaning, it's unlikely that their use will have rational effects. Sherman and Austin first learned to classify three foods and three tools. Their classification entails simply placing an item if it was a food in one bin, if a tool, then it should be placed in another bin. They next learned to categorize not the foods and tools per se, but rather photographs of them. And then they learned to label the lexigrams of each of these six training objects through the use of two new lexigrams, one for food and one for tool. In other words, they were presented the lexigram for bean cake, or for orange, or for screwdriver, or for straw. And on the basis of looking at the lexigram, labeled it as either a food or a tool through use of two other lexigrams, one for food, one for tool. In the final test, Sherman and Austin were presented for the first time in this experiment with 17 previously learned lexigrams that stood for a variety of foods and drinks and other tools that they had learned to work with across the months before. The question was whether or not they could accurately label each one in turn into the correct category through use of two lexigrams, one for food and one for tool. The answer was yes, they could. Only one error was made, and that was Sherman calling a sponge a food rather than a tool. This might not have been an error from his perspective, however, for he literally consumed sponges as he sucked on them avidly when they were soaked with favorite juices. He used the sponges to get juices otherwise out of reach. Sherman and Austin also demonstrated impressive symbol-based cross-modal matching. They were able to look at a lexigram, then reach into a box into which they could not see, feel around, and then from the assortment of objects in that box select the one represented by the lexigram. They did this by touch alone without specific training to do so. They could do this because the lexigrams were meaningful to them and served to represent items not necessarily present in either time or space. Apes have the capacity for entry-level semantics, that is the capacity to use and to understand symbols that represent events, actions, and things not necessarily present. Sherman and Austin clearly showed in several tests that they had come to think in terms of symbols, as in the example I just gave you, and to make decisions based on them. They could also go to a room by themselves, look at an ever-changing, randomly selected array of foods and tools, I'm sorry, of foods and drinks, and decide which one they would like to have, go back to a keyboard, announce which one it was that was their choice, go back to that room by themselves and with about 90% reliability would return with the item which they had announced that they would get. The 20th century will be noted for a wide variety of scientific and technological advances, including a powered flight, antibiotics, space travel, and the breaking of the genetic code. It also should be noted as the century in which major psychological continuities have been documented between animal and human life. We've accepted the biological continuity for a long time, but the psychological continuity data are now strong enough that we should admit that, yes, that tie, too, is in place. Charles Darwin, 1859, was quite right when he anticipated continuity in mental processes, some of which provide for language. Though none will argue that any animal has the full capacity for humans for language, evidence is that at least some of them have very impressive competencies for language. Again, I encourage you to attend the lecture tomorrow when the sentence comprehension of Kanzi is explored in greater detail by Sue Savie Drumbau. Traces of language are sufficient to our point that an ape is unlikely ever to be invited to the Nobel conference is beside the point. Research with apes serves to encourage a reconsideration of behaviorism. Our research indicates that learning can be much, much more than either Pavlovian, respondent, or Scaniaean, opulent conditioning. It is clear that primates interrelate their learning experiences that have been obtained in various ways, in various contexts, in various times. So both respondent, conditioning, a la Pavlov, and the operative behaviors, a la Skinner, clearly are important to survival. In addition, however, we propose that they provide the foundations for still other processes, still other kinds of behaviors that operate silently in the sense that we might not even know that something new is being understood or put into place. We call these emergence. Emergence are competencies for new kinds of responses, and they include, I believe, the phenomenon known as stimulus equivalence for those of you who follow Murray Sidman's distinguished work. Emergence are dependent upon, by and large, a highly complex brain and appropriate early experiences, as we shall see here in a moment. Emergence include language and counting and concept formation and provide for generalized competencies of adaptation. Thus, emergence differ from respondents and operants in their origins, properties, and functions. They are not reflections of specific reinforcement of specific responses to the stimuli, though that might contribute information and the engine for their formation. And to the degree that we have a creature with a highly complex brain, such as the chimpanzee and human, we're more likely to observe these emergent behaviors than we are in simpler forms of life with simpler brains. The other side of the coin is that if with a creature with a highly complex brain, we deny it an experience appropriate to the complexity of its brain, we will likely compromise for the rest of that primate's life, its ability to learn. Here we have two curves for two groups of chimpanzees studied decades back at the Yorkies Primate Center. Along the baseline, it is clear that these two groups could learn simple discriminations with equal facility. What were these groups? Well, the curve at the top was for a group of eight chimpanzees that were wild-born and group-reared from the very beginning of the study, very shortly after their birth. For the other six and the other two weren't even tested, the other six in the restricted group were reared in impoverished environments for the first two years of their life. At the end of two years, they were group-reared with those chimpanzees, and at the age of 14 and 15, we tested them with the ability to transfer. As you see, the restricted chimpanzees were basically a population apart. Those are range markers, not standard deviations. We have, then, I believe, emergence occurring very strongly as a function of the complexity of the organism and the kind of treatment condition at early ages are particularly powerful. And we know that chimpanzees are much more compromised by impoverished early rearing than are rhesus. The literature teaches us that very clearly. But, on the other hand, as we've made clear, I trust, rearing chimpanzees in language-structured environments can result in their acquiring even language comprehension, the comprehension of speech. And Kanzi's production of language is at the level of a year-and-a-half-old child through use of gestures and his keyboard. Let's take two specific examples to try to make what we mean by emergence more clear at a more concrete level. Austin has always liked mirrors. And he's always liked to look down his throat. And when he discovered that he could play to a TV camera and see his mouth on the monitor, he was really smitten by this. But, for those of you who have tried to look down your throat, it's a rather dark place. Austin seemingly concluded that as well. What was his solution? He went and got a flashlight and did his best to shine that flashlight into his mouth so as to enhance the quality of the picture. He's taken from video as indeed he did that very thing. A second example, a point here on emergence is that Austin was never trained to do this. But, nonetheless, he knew what a flashlight was, he knew what he looked like, and he knew how to put the two and two together in a new situation with a TV monitor so as to enhance the outcome of what he was interested in. A second example entails a rhesus monkey that we trained to use a foot to activate the joystick on the monitor. And here at this point is that skills emerge later on that are not necessarily manifested during training. So, here in the schematic please note that the monkey had to hang by his hands in order to activate the computer and then it had only a foot and it could use either foot to learn to use that cursor to chase the moving target on the screen. When it got good at that and only when it got good at it then was he extended the opportunity for the first time to use a hand. In other words, he was able to use either his right or left hand. What was the contribution, if any, what was the effect of any of learning with his foot upon his ability to use his hand for that same task? I ask a number of my colleagues about that and their predictions ranged anywhere from it won't help them, him at all, not at all. Nobody thought that it would make it tougher but it might not help at all. Others thought well it might help 30-40% some 80%. But the exciting thing was that everybody was wrong. Everybody was wrong. This monkey showed right off that he was better with his hand than he had ever been with his foot. He didn't learn how to use his foot alone to do that task. He learned the task and how to use it, use even the more facile movements of his arm and hand to do it. So the function of emergent processes is similar to that of speciation. Emergence are the fountain head of behavior. New ways of doing things, of creating better solutions to old problems and ingenious solutions to new ones in an efficient and timely manner. Thus, the emergent framework can help us better understand how rearing environments can either foster or deter language and even alter the kind and quality of the learning processes. Rhesus, with a lot of experience in that computerized test system, finally come to manifest learning processes otherwise reserved for the apes. And now we begin to look to the future and now we also end where we began with Descartes and the ringing of Albert Screaming in our ears. Contrary to the historical doctrines of Descartes and others of centuries past, apes are not unmindful creatures. Other speakers at this meeting will make that abundantly clear. Neither they nor we are beast machines and we should and must do better by for and with the apes and animals generally. With them we must learn to share this planet lest all is lost. We must do better much better at conserving nature the only adequate home for apes and their future generations. It will be with the future of nature and the great apes that history will declare eventually for better or worse the destiny that will be ours and that destiny will be one and the same as for the great apes. What we let happen to the apes will eventually happen to us and it is ever so important that that destiny be a successful one. Thank you. Card right here. Is that being given out? Are there cards back there? Ladies and gentlemen, we'll take about 15 minutes for questions and responses 20 minutes. I believe that there are cards available to you in which you can write a question to Dr. Rumba. Are they coming down the side aisles? Do you see them? I hope you see them. I can't see them. I have these lights in my eyes. If you'd raise your hand the ushers will give you a card and all of these cards will go to Dr. Rumba whether they are read by Dr. Brown or not. We'll select three or four of them five of them from you to help us with our conversation. If this is your first Nobel conference let me tell you that sometimes some of the most exciting moments are the synergy that the question and answer that takes place between the lecturer and between the audience and the lecturer. So please do ask your questions on the cards and pass them up. First of all, are there any members of the panel who would like to either respond to Dr. Rumba or ask a question? Dwayne, what you've presented is a picture that seems to be a little bit different from one that has been made popular recently by Steven Pinker in his book, The Language the reason that humans are able to use language is because natural selection has specifically shaped linguistic abilities. Can you explain how your position is different from Pinker's? What you seem to be saying is that the brain has emergent properties that happen to produce linguistic abilities. Do we then go on and say that the reason that the human brain can use language so effectively is simply because it has more emergent properties without any action of natural selection? One must first of all have an attribute present before it can be selected for and it will be selected for to the degree that it has can make a significant contribution to survival and reproductive success. The human brain has come to be three times as large as that of the great apes and thus it would stand to reason that we would have more brain power more computational ability to turn over to try to integrate all of the various perceptions and experiences that we have. The chimpanzees that the bonobos competencies for language do not begin to challenge ours in terms of finesse in terms of excellence but not outstanding. We now have in the database of ape language research of the last 30 years sufficient evidence to say that these creatures are capable of learning the meanings of abstract symbols and coming to think in terms of them and solve problems. Now that being selected for in the evolution of humans and all of this goes back to an older root as you know would in fact allow for possibility that some fortuitous events might allow for more pressure to produce a larger brain. Now we have clearly as humans the capability for speech and hearing that is remarkable. It is through speech that we have the most facile means of communicating language although parrots can quotes talk quotes they can't really talk as we do but notwithstanding they can learn the meanings of symbols. Speech and hearing and hearing systems specifically for speech came into place certainly did allow for the opportunity to count more and more upon linguistic processes linguistic products which the great apes in fact I don't think generally have among themselves. In the field they might be using language in ways that have yet to be discerned but in captivity it isn't like they have exploited the hilt out of the language skills that they have learned but that said they do very efficiently and effectively and appropriately use their language skills and that's the best way to access things. So I would I would take issue with Pinker language is a continuum the great apes very definitely have a place on that continuum we're further along on that continuum these reflect the common route in the past. We have a question from the audience how much progress has been made in understanding the natural language of apes that is can we use their languages as well as teaching them to use ours. I'm going to defer on that question and tell tomorrow for Dr. Sue Savie Drumbach she has been in Wamba's IAIR and has some ideas and materials to present that pertain to that I think it's clear that the apes have probably learned a bit more about our language than what we have learned of theirs. That's a fair statement. Here's another question. Do apes ever teach others what they've learned can they pass on what they learn from generation to generation? Chimpanzees and bonobos gorillas and orangutans are to my best information not noted for being teachers we humans are the teaching species but that said the mothers do teach their infant some things like where not to go don't do this they try to keep them in safety but they don't sit down generally and try to instruct their infant how to do a specific thing though there are some instances reported as from the fouts and even from our own laboratory there's some but the apes are not dedicated teachers as are humans and as the faculty of Gustavus. This question deals with operationalization I guess. Did Lana's lexagrams include which is as well as nouns and adjectives such as banana and green or did you infer the connective language? She had a lexagram that was the way we glossed it, the way the computer glossed it and the way it was printed out was it was literally spelled out as that's that is it was one lexagram but yes there was a lexagram for that just as there was for each color and her foods and she used the word please incidentally to signal the computer or at least the computer took it as a signal that a request was being formulated we weren't really requiring of her that she be polite it's just that we decided to call that first word please if indeed a request was being formulated similarly we use a period key not to require of her punctuation skills but rather the computer program needed some kind of a signal that she had completed and needed to and the computer then evaluated it for appropriateness. Other members of the panel Dr. Kaufman Is there anything analogous to the asking of questions by apes? Can they formulate a question? They do they do very clearly indicate that they have questions about what it is that's going on what it is that's going to happen next and we take the question do they formulate questions formally with Lana and I'm not saying she was fully cognizant of what she was doing but we do have recorded episodes where for example she wanted a container for which she had no name and she would go through her list of bowl, box, can and shoe and so on and it still wouldn't be appropriate and then she did on two occasions ask what name of this and Tim Gill said like bowl name of this that was before she had learned the name bowl whereupon then she said yes and then said please give this bowl to Lana so it's again I'm sure that in the presentation of my wife and colleague you'll see rich examples of how otherwise chimpanzees ask questions through their gestures through their glances and how they give verbal assurance vocal assurance with their answers when indeed it's something that they've been wanting they'll give a very excited wow, wow, wow, wow Other members of the panel Dr. DeWall I want to add just a little bit about the question of whether they pass on knowledge to each other because I'm sure Richard can say a few things about that also is that there is I think a tremendous amount of knowledge being passed on from older generations to younger generations or from one individual to another but teaching is not as you say teaching is not a prominent feature of that but there are many mechanisms in which this may happen imitation is one is heavily contested there's a very controversial issue at the moment whether apes have imitations or not but recent studies even on hamsters now are indicating imitation so I'm not surprised that chimps have it and there are other ways of it's called stimulus enhancement they see where other individuals find something and they go to the same spots and find the same things or use the same tools maybe not exactly imitating tool use but using at least the same instruments that others are using so I think there's a tremendous flow of knowledge from one individual to another and you can see also in the field and also in captivity some groups have certain amounts of knowledge that other groups don't have and so you see intergroup differences that we can only explain as a form of traditional learning that's going on so there's a lot of that teaching is not part of it seems not a big part of it at least but there's a lot of transmission of knowledge going on absolutely the offspring, the young learn a great deal by observing what their significant others if you will their mothers and friends do in the situation and you'll hear a lot about that both from you and also from Ted Searle another question from the audience you left me with the impression that you felt psychological response did not have a biological basis am I correct no you're not correct I believe very very much that there is a very tight tie very tight link between all of psychological processing all of behavior and biological functioning which in turn is shaped by genetics and early environment and what we eat and all kinds of ambient conditions but it is the case that some biology of some life forms comes forth with new emergent capabilities, surprises and clever in ways that you can't really account for by the standard behavioristic frame of thinking biology is very definitely there and we will be doing some studies of brain imaging to see what parts of the brain are activated with various kinds of these language functions and other non-language functions while we're on behaviorism I'll ask a question I did my doctoral studies at a school where B.F. Skinner had been some decades before and there were still romantic stories about research assistance being sent out to the grain elevators for pigeons which were his specimens if B.F. Skinner had studied chimps instead of pigeons might our view be different I believe so I believe so I really believe that B.F. Skinner would have benefitted a great deal from working with chimpanzees he did visit our laboratory and was profoundly impressed with the richness of the behavior of the animals far beyond anything that he had expected I will say however there was one psychologist, Kenneth Spence who spent time at the Yorkies laboratories when they were in Florida and I couldn't see that he was positively impacted upon by any kind of that experience B.F. Skinner made great contributions but we need to build upon those contributions and those of other behaviors as we try to better understand what it is to be alive to be competent to do things as living creatures rather than just as particles turn again a question from the audience from the psychological to the physiological the audience member writes considering evolution why is it that chimpanzees DNA is closer to humans than gorillas when in evolution the closest primate to humans is the gorilla or ape I don't know can anyone on the panel answer that I don't understand the question again I'll read it again considering evolution why is it that chimpanzees DNA is closer to humans than gorillas when in evolution the closest primate to humans is the gorilla or the ape well chimps are apes chimps are apes just like gorillas and so we can draw a quick tree but it's assumed that the split between us and the gorilla is like 8 million years ago and between us and the pan bronze which is bonobos and chimps is like 6 million years ago so the split is a bit earlier so they're a bit more distant yeah the question seems to be based on a false premise which is that humans and gorillas are most close to related apes and not it's the humans to the chimpanzee bonobo line I will briefly discuss this tomorrow we have one more question from the audience have the apes been able to teach you or us anything have we as humans let ourselves be captured and tested well that's a very interesting question it is true that behavior is very much based in social contexts I am always amazed as to how apes behave differently around different people I know that when Sue goes into one of the areas the even the common chimpanzees that don't receive so much attention from her now come to life in a way that they don't for anybody else they just don't for anybody else it isn't that Sue or we have been captured by them it is rather that by working with them in a very close social way essentially from birth there are behaviors there are learning experiences for the chimpanzee for the binobo and also for the human which allows for an escalating graduated kind of richness productivity commerce so in a sense the apes change the people and the people change the apes but the merit of science is that you can assess what that is in at least some special situations such as asking a chimpanzee in a blind test situation where he couldn't see Sue to do many things that he had never been asked to do and that he could do that tells us that indeed he's learned things and as he does those things we know that we've learned things about him that we never would have known before Thank you Dr. Rumbaugh for your stimulating lecture thank you members of the panel we will reconvene at 1.30 this afternoon and we'll see you in Oak West Lake