CAMBRIDGE, Mass. — Seated in a cheerfully cramped monitoring room at the Harvard University Laboratory for Developmental Studies, Elizabeth S. Spelke, a professor of psychology and a pre-eminent researcher of the basic ingredient list from which all human knowledge is constructed, looked on expectantly as her students prepared a boisterous 8-month-old girl with dark curly hair for the onerous task of watching cartoons.


The video clips featured simple Keith Haring-type characters jumping, sliding and dancing from one group to another. The researchers’ objective, as with half a dozen similar projects under way in the lab, was to explore what infants understand about social groups and social expectations.


Yet even before the recording began, the 15-pound research subject made plain the scope of her social brain. She tracked conversations, stared at newcomers and burned off adult corneas with the brilliance of her smile. Dr. Spelke, who first came to prominence by delineating how infants learn about objects, numbers, the lay of the land, shook her head in self-mocking astonishment.


“Why did it take me 30 years to start studying this?” she said. “All this time I’ve been giving infants objects to hold, or spinning them around in a room to see how they navigate, when what they really wanted to do was engage with other people!”


Dr. Spelke, 62, is tall and slim, and parts her long hair down the middle, like a college student. She dresses casually, in a corduroy jumper or a cardigan and slacks, and when she talks, she pitches forward and plants forearms on thighs, hands clasped, seeming both deeply engaged and ready to bolt. The lab she founded with her colleague Susan Carey is strewed with toys and festooned with children’s T-shirts, but the Elmo atmospherics belie both the lab’s seriousness of purpose and Dr. Spelke’s towering reputation among her peers in cognitive psychology.


“When people ask Liz, ‘What do you do?’ she tells them, ‘I study babies,’ ” said Steven Pinker, a fellow Harvard professor and the author of “The Better Angels of Our Nature,” among other books. “That’s endearingly self-deprecating, but she sells herself short.”


What Dr. Spelke is really doing, he said, is what Descartes, Kant and Locke tried to do. “She is trying to identify the bedrock categories of human knowledge. She is asking, ‘What is number, space, agency, and how does knowledge in each category develop from its minimal state?’ ”


Dr. Spelke studies babies not because they’re cute but because they’re root. “I’ve always been fascinated by questions about human cognition and the organization of the human mind,” she said, “and why we’re good at some tasks and bad at others.”


But the adult mind is far too complicated, Dr. Spelke said, “too stuffed full of facts” to make sense of it. In her view, the best way to determine what, if anything, humans are born knowing, is to go straight to the source, and consult the recently born.


Decoding Infants’ Gaze


Dr. Spelke is a pioneer in the use of the infant gaze as a key to the infant mind — that is, identifying the inherent expectations of babies as young as a week or two by measuring how long they stare at a scene in which those presumptions are upended or unmet. “More than any scientist I know, Liz combines theoretical acumen with experimental genius,” Dr. Carey said. Nancy Kanwisher, a neuroscientist at M.I.T., put it this way: “Liz developed the infant gaze idea into a powerful experimental paradigm that radically changed our view of infant cognition.”


Here, according to the Spelke lab, are some of the things that babies know, generally before the age of 1:


They know what an object is: a discrete physical unit in which all sides move roughly as one, and with some independence from other objects.


“If I reach for a corner of a book and grasp it, I expect the rest of the book to come with me, but not a chunk of the table,” said Phil Kellman, Dr. Spelke’s first graduate student, now at the University of California, Los Angeles.


A baby has the same expectation. If you show the baby a trick sequence in which a rod that appears to be solid moves back and forth behind another object, the baby will gape in astonishment when that object is removed and the rod turns out to be two fragments.


“The visual system comes equipped to partition a scene into functional units we need to know about for survival,” Dr. Kellman said. Wondering whether your bag of four oranges puts you over the limit for the supermarket express lane? A baby would say, “You pick up the bag, the parts hang together, that makes it one item, so please get in line.”


Babies know, too, that objects can’t go through solid boundaries or occupy the same position as other objects, and that objects generally travel through space in a continuous trajectory. If you claimed to have invented a transporter device like the one in “Star Trek,” a baby would scoff.


Babies are born accountants. They can estimate quantities and distinguish between more and less. Show infants arrays of, say, 4 or 12 dots and they will match each number to an accompanying sound, looking longer at the 4 dots when they hear 4 sounds than when they hear 12 sounds, even if each of the 4 sounds is played comparatively longer. Babies also can perform a kind of addition and subtraction, anticipating the relative abundance of groups of dots that are being pushed together or pulled apart, and looking longer when the wrong number of dots appears.