Easy as 1,2,3
by Rob Shumaker
The Smithsonian National Zoos Think Tank is not a typical Washington think tank. It is a place where orang utans (Pongo pygmaeus) live and work, and reinforce the notion that the mental differences between humans and great apes are not in kind, but degree. Since cognitive research began at Think Tank in 1995, the work has focused mainly on language, using a vocabulary of abstract symbols with English syntax to explore how orang utans learn and understand language. They can now use symbols consistently from the verb, food-name, and object-name categories with greater than 90 percent accuracy.
More recently, I have turned my attention on the quantitative abilities of orang utans, evaluating quantity judgment, subitization, ordination, and countingall aspects of numerical competence. Quality judgment involves a comparison between, and subsequent judgment of, distinct quantities; subitization is the rapid labeling of small quantities that are presented simultaneously (for humans this usually tops out at five to six items); ordination is the recognition that five is more than three, and that two is less than six; counting, considered the most complex of these abilities, is the process of indicating an absolute value for a quantity and labeling it.
Chimpanzees have generally tested extremely well on their ability to discriminate between quantities of foods and choose a larger array. Moreover, chimpanzees can label quantities from zero to nine with Arabic numerals and one individual, studied by Sarah Boysen at Ohio State University, can correctly add sets of items or Arabic numerals. Interestingly, once this chimp understood how to label individual quantities with their correct number, she required no further instruction to add them. In one particularly interesting series of chimpanzee studies, Boysen created tasks in which the chimp was given a choice between two quantities of candy varying from one to six pieces; but unlike previous experiments, the quantity the chimp chose was subtracted and the remainder given as a reward.
I was anxious to recreate these experiments with orang utans Azy and Indah, cognitive research subjects at Think Tank. Each was presented with two choices and, again, choosing the smaller quantity resulted in a larger reward. In this case, the choice was grapes and they were presented with all fifteen possible paired combinations of one through six. Early in the study, Azy chose the smaller quantity only about 50 percent of the time. However, in session eight; his score jumped to 85 percent and, from then on, never dipped below 80 percent. (He was 100 percent accurate in session 21.) Indah didnt break through until session 14 when she reached 70 percent accuracy, scoring at or above 80 percent through the end of the study.
In Boysens tests, where choosing the smaller of two quantities of candy resulted in receiving a greater reward, chimpanzees chose the smaller quantity 27 percent of the time. In addition, they generally did worse in trials with a larger risk, such as receiving one candy versus six. However, in otherwise identical trials that used numerical symbols rather than candies, they were able to choose the smaller quantity 66 percent of the time. In fact, trials with candies could be alternated with trials using numerals, and average scores within each type of trial remained consistent: low for candies and high for numerals. The chimpanzees clearly understood how to choose less to get more, but simply had trouble choosing less when presented with larger physical quantities rather than numerals. When rocks were substituted for candies, they were still inclined to choose the larger quantity thus receiving less. Orang utans Azy and Indah did not have this limitation when grapes were used as a reward.
The Think Tank study showed that orang utans use a variety of cognitive skills to make the optimal choices, and their skills are comparable to adult humans. The ability to spontaneously order all possible combinations of quantities, which emerged at session eight for Azy and session 14 for Indah, relied on an understanding of ordinality, a skill humans develop by three to four years of age. Most important, the orang utans were able to work out that to get more grapes they had to choose less, and did so without any instruction or direct guidance during the process. Now, Azy and Indah are learning to use the Arabic numerals 1, 2, and 3 with intriguing results. Visit Think Tank to find more.
Rob Shumaker is a researcher at the National Zoo.
ZooGoer 31(4) 2002. Copyright 2002 Friends of the National Zoo. All rights reserved.
