New Research Finds Some Animals Know Their Cognitive Limits

One of the important questions in the field of animal and human psychology is whether this metacognitive capacity is uniquely human, or whether nonverbal, nonhuman animal species have a level of metacognition that approaches that of humans. Animals could demonstrate a capacity for metacognition if they could report their uncertainty or doubt when confronted with a difficult trial or situation. However, research in this area has been slow to emerge because it is inherently difficult to ask nonverbal animals whether they know, or feel uncertain, or have doubts.

Steps toward solving this problem now have been made by a research team led by John David Smith, Ph.D., associate professor in the Department of Psychology in the College of Arts and Sciences at the University at Buffalo and UB's Center for Cognitive Science. The research team includes Wendy E. Shields, Ph.D., of the Department of Psychology, University of Montana, and David A. Washburn, Ph.D., of the Language Research Center at Georgia State University.

Their research, "The Comparative Psychology of Uncertainty Monitoring and Metacognition," will be presented in the December issue of The Behavioral and Brain Sciences, one of the premier journals in the field of cognitive science.

The article describes three studies by the authors with humans, a group of Rhesus monkeys and one bottlenose dolphin that used behavioral, nonverbal measures of metacognition. In these tasks, animals experienced a mix of "hard" and "easy" perceptual or memory trials. If they completed the trial, the subjects earned a reward when correct or a timeout period when wrong, Smith says.

"The key innovation in this research also was to grant animals an 'uncertain' response so that they could decline to complete any trials of their choosing," Smith says. "Given this option, animals might choose to complete trials when they are confident they know, but decline them when they feel something like uncertainty. To show this behavioral pattern, though, animals would have to monitor some psychological signal of confidence or uncertainty and respond adaptively to it."

The researchers have shown that the monkeys and the dolphin used the "uncertain" response in a pattern that is essentially identical to the pattern with which uncertain humans use it.

Indeed, Smith says, "the patterns of results produced by humans and animals provide some of the closest human-animal similarities in performance ever reported in the comparative literature."

Moreover, it is clear that a higher-level cognitive interpretation of the results is warranted -- low-level behavioral explanations cannot explain the phenomena. In short, Smith says, "the results suggest that some animals have functional features of, or parallels to, human conscious metacognition."

They apparently know when they know and when they don't know, he adds.

Another intriguing finding emerging from this area of research is that species that are less cognitively sophisticated (e.g., rats and pigeons) have not thus far expressed the same capacity for cognitive monitoring or cognitive self-awareness as that expressed by the monkeys and dolphin in the studies. Smith and his co-researchers point out that by using the same metacognitive paradigms broadly across species, scientists may be able to draw the map showing which species have evolved cognitive self-awareness. This could reveal when in evolution reflective cognition emerged and how widespread this capacity is among animals.

An important feature of the publication of the article is that 21 commentaries on the research by many of the world's most distinguished scientists in the areas including comparative psychology, developmental psychology, human cognition and philosophy will be published simultaneously. A detailed response to these commentaries by the researchers also will be published in the same issue of The Behavioral and Brain Sciences.

Together, the original article, the commentaries and the response inaugurate a new area of comparative cognition research and offer a rare glimpse at the broad, interdisciplinary discussion that attends the advent of a new research area.

A shorter summary of the research will appear later in an issue of the journal Current Directions in Psychological Science, another highly influential psychological journal that showcases important new developments.