Additional Studies Into Ape Language And Primate Intelligence

The Chantek Project:

Bringing A New Perspective To Orangutan Intelligence

Soon after receiving her Ph.D. in anthropology, Dr. Lyn Miles was given the opportunity to attempt to train an orangutan in pidgin American Sign Language. Chantek, a six-month old orangutan from the Yerkes Laboratory of Primate Biology, became Lyn’s student in 1978 at the University of Tennessee at Chattanooga.

During his eight years at Tennessee, Chantek learned a total of about 150 signs. According to Miles, the language training was broken down into three stages: (1) establishing the rules for human communication and discourse; (2) teaching specific gestural signs; and (3) encouraging his independent use of signs to transmit meaningful information about himself and his environment. Miles noted that Chantek used signs for deception, indicating a more sophisticated cognition than the use of signs for straightforward requests and comments. For example, he often signed “cat” when his caretakers saw no animal, apparently to distract attention from testing or to avoid returning to the trailer when on an outing. Chantek also provided Miles with many examples of apparent imitation. One example was his use of the sign for imitation, “do-same,” which he had never been taught. He used the sign to reverse the roles of himself and a caretaker. The caretaker had been asking Chantek to imitate various actions, when Chantek very carefully and correctly signed “do-same” while maintaining eye contact with the caretaker.

The Chantek project destroyed the common assumption that the slow and deliberate orangutan was also the stupid orangutan, relative to the other great apes. Chantek developed linguistic skills in a manner strikingly parallel to the way deaf human children and other great apes learn sign language. Thus Chantek added to the evidence that the nonhuman apes share with humans significant cognitive underpinnings for the comprehension and production of language. Finally, Chantek’s ability to imitate actions, to be deceptive, to be creative and to be self aware, makes him, like Koko, Kanzi, and many other language-trained apes, appear to be almost human.

In addition to his intelligence, Chantek consistently demonstrated self-awareness. To test this, researchers place a mark on an animal’s forehead when they are anesthetized or otherwise unaware. When alert, the animal is given a mirror. If the animal immediately touches the mark, the test is passed. Animals that do not have self-awareness respond to their image as though it were another individual. Some display at the image. Others reach behind the mirror. Miles tested Chantek in this accepted method, watching his development of self-awareness until it became consistent at age 3 1/2. After that time, Chantek actively sought out mirrors in order to groom himself. Contrary to earlier claims, bonobos, chimpanzees, gorillas and orangutans that have received language training all present clear evidence of self-awareness.

Orangutan Language Program

Dr. Robert Shumaker at the Think Tank exhibit at Smithsonian National Zoological Park, and later at Great Ape Trust, explored the abilities of orangutans to use symbols and syntax to express their thoughts. The orangutans learned to use a symbol-based language that is presented on a computer monitor. The monitor screen had large "buttons" big enough for orangutan fingers.

The dictionary available on the monitor contained about 70 symbols. All symbols were abstract and had no visual relation to what they represented. There were seven categories of symbols, each containing ten individual symbols. The categories were: foods, non-food objects, proper names of people, proper names of orangutans, verbs, adjectives and arabic numbers.

Each category of symbols had its own specific exterior shape. For example, a rectangle alone meanty "food" and a circle alone meant "non-food object." Individually, the interior components of each symbol were meaningless. It was the arrangement within the exterior shape that gave each symbol a specific meaning. In addition to the major categories, there were symbols that meant "send," "clear," "yes/good," and "no/wrong." The dictionary was able to be expanded as the orangutans learned more symbols.

In addition to the language project, additional research focused on the acquisition, comprehension, and demonstration of numerical competencies in orangutans. Specific cognitive skills that were investigated included quantity judgment, labeling of discrete quantities with Arabic numerals, and representation of the numerical abilities by means of the development of a token economy.

The Ai Project

In November 1977, Ai, a one-year-old female chimpanzee, became the first subject of an ape language research project at the Primate Research Institute of Kyoto University (KUPRI), Japan. The goal of the project was to study the acquisition process of an artificial language and the corresponding brain mechanisms. Adopting the computer-controlled lexigram developed by Duane Rumbaugh and his colleagues, KUPRI’s Tetsuro Matsuzawa hoped to investigate how specific visual symbols such as those representing the names of individuals, objects, and actions, could be established in chimpanzees.

Ai participated in her first computer task alone inside an experimental booth in April 1978. The original keyboard contained three panels. Each panel consisted of 35 keys arranged in seven rows and five columns. Touches to back lit keys produced a feedback sound, the key light faded, and facsimile of the lexigram appeared above the keyboard. There was a window through which the tester showed an object to the chimpanzee, who sat voluntarily on the bench facing the apparatus. After gathering data on color perception, shape perception and visual acuity, Matsuzawa and other researchers proceeded to study the recognition of numbers. Ai became the first chimpanzee to use Arabic numbers to represent quantities.

Based on the results of the initial studies, the Ai Project went on to look at several other areas, including short-term memory, learning sequences, face recognition, perception of moving images, perception of visual illusions, representation of symbols, categorization, and computer-assisted drawing. The work of Matsuzawa continues to offer a unique window into the perceptual and cognitive world of the chimpanzee.

In addition to the language studies with Ai, Matsuzawa began a field experiment in Japan to study the tool-using behavior of chimpanzees. This study has revealed that a form of observational learning plays a key role in the transmission of knowledge and skills from one generation to the next. The study also showed the importance of social relationships in learning.

Between 1982 and 1983, Matsuzawa had the unique opportunity to raise a chimpanzee baby at home, and to compare the infant with his own. Through his observations during these years and his field experiments, Matsuzawa recognized the importance community plays in the cognitive development of infant chimpanzees. Bearing this in mind, KUPRI developed a new paradigm for studying cognitive development in chimpanzees. It is based on a relationship between a mother chimpanzee, an infant chimpanzee and a human tester. Based on the work completed to date in the Ai project, Matsuzawa has established one of the most accurate representations within the long debated field of human-chimpanzee developmental parallels and contrasts.