Strange as it might seem, not all animals can immediately recognize themselves in a mirror. Great apes, dolphins, Asian elephants, and Eurasian magpies can do this—as can human kids around age 2. Now, some scientists are welcoming another creature to this exclusive club: carefully trained rhesus monkeys. The findings suggest that with time and teaching, other animals can learn how mirrors work, and thus learn to recognize themselves—a key test of cognition.
“It’s a really interesting paper because it shows not only what the monkeys can’t do, but what it takes for them to succeed,” says Diana Reiss, a cognitive psychologist at Hunter College in New York City, who has given the test to dolphins and Asian elephants in other experiments.
The mirror self-recognition test (MSR) is revered as a means of testing self-awareness. A scientist places a colored, odorless mark on an animal where it can’t see it, usually the head or shoulder. If the animal looks in the mirror and spontaneously rubs the mark, it passes the exam. Successful species are said to understand the concept of “self” versus “other.”
But some researchers wonder whether failure is simply a sign that the exam itself is inadequate, perhaps because some animals can’t understand how mirrors work. Some animals—like rhesus monkeys, dogs, and pigs—don’t recognize themselves in mirrors, but can use them to find food. That discrepancy puzzled Mu-ming Poo, a neurobiologist at the Shanghai Institutes for Biological Sciences in China, and one of the study’s authors. “There must be some transition between that simple mirror use and recognizing yourself,” he says.
So Poo and his colleagues put three young male rhesus monkeys through an intensive training program. Each monkey was secured in a chair facing a mirror, and researchers flashed a red laser pointer at random positions nearby. When the monkeys touched the dot, they received a treat. Sometimes, they could see the dot only by using the mirror. “The monkey has to learn that the hand in the mirror is his own hand. And he has to learn how to control it precisely by watching it in the mirror,” Poo says. “That’s the key.” Three other monkeys, who served as controls, weren’t trained to respond.
After several weeks of training, the scientists gave each monkey a classic MSR test, applying an odorless red, black, or green splotch to the animal’s cheek or eyebrow and putting it in a cage with a full-length mirror. The trained monkeys didn’t hesitate; they looked in their mirrors, pawed at the marks, and smelled their fingers, the scientists report today in the Proceedings of the National Academy of Sciences. When left alone and unmarked, they did just what you’d expect from someone who’s just discovered how mirrors work: They bent over and inspected their genitals, opened their mouths to examine their teeth, and carefully combed their hair. “That showed that they recognize themselves and know they are looking at their own bodies,” Poo says. The control monkeys did none of these things.
“The training seems to switch on a light bulb in the monkeys,” says Reiss, adding that even animals in the self-awareness club have to learn about mirrors. “[The spontaneous users] learned on their own what the monkeys learned from their training.”
The study suggests that the mental processes underlying the mirror test are shared across primates, but that monkeys need more information to pass it, says primatologist Frans de Waal of Emory University in Atlanta, who was not involved with the study. Still, he and others point out that untrained monkeys continue to fail the test.
“These are not spontaneous, self-directed behaviors, which is the point of the MSR test,” says Lori Marino, a biopsychologist and executive director of The Kimmela Center for Animal Advocacy in Kanab, Utah, who was not involved in the study. She thinks the study has “in no way challenged [the test’s] validity.”
Next, Poo and his colleagues plan to scan the brains of monkeys during their mirror training to see which brain circuits might be changing in the process. Meanwhile, Poo urges other scientists to come up with new tests for self-recognition in other species–a skill he suspects may be more common than we think.