Spotlight: Miller

What makes you so sure I'm not a cat?

In daily conversation, "learning" means something that happens in classrooms, while "memory" refers to misty recollections of times gone by, or desperate efforts to recall the location of your keys. But much of the brain's work in learning and memory is something else altogether, something much more central to our ability to make sense of the world. The ceaseless human urge to categorize everything we encounter—the ability to learn what properties define a dog, for instance, to apply that rule instantly and unconsciously whenever we need to, and to repeat this process with millions of other patterns of data—is a quiet, monumental feat of learning and memory.

Without this ability to assign categories of sameness and difference, the world would present an endless, overwhelming cascade of surprises; every coffee cup or paperclip would require individual investigation. Yet despite their central role in what we consider intelligence, until now little has been known about how these categories take shape in the brain.

Focusing on the prefrontal cortex, a region highly developed in humans and other primates, Earl Miller's laboratory has made several breakthroughs in understanding how these categories are created and stored. In one experiment, monkeys who had never seen a real cat or dog were trained to recognize images of various cats as being "cats," and images of various dogs as "dogs." Later, when presented with "morphed" images that blended features of cats and dogs, the monkeys reliably recognized anything more than 50 percent catlike as a cat, and vice versa. The Miller lab was able to show for the first time that these learned concepts are actually stored in individual neurons. More recently, the lab has also demonstrated that while monkeys lack the human ability to "count" with language, they can nevertheless visually distinguish numbers of objects up to five—and that this ability, too, appears to be stored in individual cells in the prefrontal cortex, specialized to respond to specific numbers.

Marking a crucial step towards deciphering how the brain processes abstract concepts, such discoveries could eventually lead, for example, to new strategies for teaching arithmetic to children.