Sea lion makes waves with logical leaps.

If psychologists soon credit many animals with markedly greater mental dexterity, you can blame it on Rio -- not the city, the sea lion. Although not trained to do so, Rio makes logical judgements of a type formerly thought possible only by humans.

Rio's feat seems fairly simple. She learned that pairs of objects go together -- say, a ring and a baseball bat, and the same baseball bat and a clothes hanger--and then realized on her own that the ring, bat, and hanger form a group of interchangeable objects. Thus, if she saw the hanger, she knew it belonged with the ring because both of those items belonged with the bat.

Scientists refer to this as the ability to form an equivalence class. Categorization of this type depends on making the logical assumption that if A equals B and B equals C, then A equals C.

Over the past 20 years, numerous investigations have failed to uncover evidence of equivalence thinking among chimpanzees, monkeys, pigeons, and other nonhuman animals. Psychologists often assume that language allows humans to devise equivalence classes.

Rio's performance challenges that theory, asserts psychologist Ronald Schusterman of the University of California, Santa Cruz. However, equivalence concepts may represent a prerequisite for learning language, he argues.

Schusterman presented his findings at the First International Congress on Behaviorism and the Sciences of Behavior, convened in Guadalajara, Mexico,

earlier this month. Researchers familiar with the study endorse his conclusion, with varying degrees of confidence.

Unlike other scientists investigating equivalence thinking in animals, Schusterman exposed Rio to training and testing that gradually increased in complexity. He devised 30 equivalence classes, each consisting of three different objects presented on large signs. For each class, Rio first learned that choose object B (the baseball bat) upon seeing object A. At that point, Schusterman presented object B first and gave Rio several choices for a match; she correctly chose object A in eight of 12 trials.

He then repeated the same training process with object B and object C (in this case, a hanger) for each equivalence class. When shown object C first, Rio correctly chose object B on 11 of 12 trials.

Finally, Rio demonstrated equivalence thinking. She matched the appropriate object C with object A on all 12 trials presented by Schusterman, although she had not been trained to do so. She also reversed this skill, matching the appropriate object A with object C on 17 of 18 trials.

"I'm quite excited about Schusterman's findings," says psychologist Murray Sidman of the New England Center for Autism in Southborough, Mass. Training in the relations between pairs of objects prior to equivalence testing appears critical to Rio's success, Sidman notes. In studies of monkeys and baboons that did not include such training, Sidman has found no evidence of equivalence thinking.

Schusterman's study "is really quite impressive," remarks psychologist William K. Estes of Harvard University. Rio displays a surprisingly agile mind, "but that doesn't mean sea lions think just like humans do," he cautions.

Schusterman's results coincide with increasing evidence that children begin to develop equivalence classes within the first few years of life, before extensive experience with language, Estes notes.

"Schusterman may have shown that equivalence concepts are not mediated by language," asserts psychologist Steven C. Hayes of the University of Nevada-Reno. "But I want to see this study in writing after it's gone through peer review."

At that point, researchers may begin using Schusterman's technique to test for equivalence classes in a variety of animals, according to Hayes.

Sea lions and other animals that live in social groups probably identify family members and neighbors by using a variety of sensory cues that make up equivalence classes, Schusterman theorizes.