As individuals, army ants have almost no brain to speak of, just a clump of neurons inside their tiny heads.

Working as a group, however, they rule the Amazon jungles, marching in formation over acres of land and flushing out thousands of insects, even scorpions, that are their prey. The ants move out and then file back in orderly lines, with the returning parties efficiently forming lanes inside the outgoing ants.

Iain Couzin, a biologist at Princeton University, has watched army ants in Panama's rain forest and figured out how they do it.

Army ants don't follow a leader. Rather, each ant acts on instinct, obeying certain inborn rules that even an ant brain can handle. Together, these simple behaviors give rise to complex-looking collective patterns, Couzin said.

Similar complex group behavior, he said, shows up in many other simple-minded creatures - herds of wildebeests, flocks of starlings, schools of fish, swarms of locusts.

"Humans do it too," said Couzin, 29, who came to Princeton from Scotland, where he trained in biology and animal behavior. "Think of crossing the street in a crowded city," he said. People are trying to go both ways across a crosswalk. You probably don't give it much thought, but you tend to get in line behind someone going in your direction, he said. From above, you can see that the crowd spontaneously forms lanes.

Scientists have tried for decades to understand organized group behavior in animals - especially the flocking of birds and schooling of fish, said Julia Parrish, a biologist at the University of Washington.

Such ordered behavior is sometimes called "self-organization" - a term that also applies to cells coming together in a developing fetus, sand grains sorting themselves, or snowflakes assembling. All are guided by similar principles.

"The magical thing about this is that you have these disordered systems that come up with elaborate patterns," Rutgers University physicist Troy Shinbrit said. "It isn't clear why they appear - there's a lot of beauty and a lot of major questions."

Self-organization doesn't require intelligence or communication skills or a leader. Each individual follows a few simple rules.

There was a big advance in this field in the 1980s, Parrish said. It came not from a scientist, but from an animator named Craig Reynolds, who created a computer simulation of a flock of birds, which he called "Boids."

The Boids, which flock on Reynolds' Web page, demonstrated a lifelike quality - organized, yet unpredictable. Other computer creatures followed.

But there was a problem with the Boids that made them an imperfect model for reality, Parrish said. The movements of individual Boids were based on some bits of knowledge that real birds wouldn't be privy to, such as how far they had flown from the center of their flock.

Real birds and fish have no clue as to the formations their groups make as a whole or where they stand within them, she said. They only know their orientation with respect to their nearest neighbors - their "local environment," she said.

Several years ago, Couzin started watching freshwater fish as they schooled in a huge tank in Canada. In a split second, the fish would switch from milling around in a group to swimming in alignment, creating a graceful ring formation.

Such acts of synchronized swimming require minimal brainpower or awareness on the part of the fish. Each individual fish need only pay attention to its distance from its neighbors. If it starts to get too close to other fish, it would be repelled. If it gets too far, it would want to edge closer.

Fish can follow those rules and remain disordered, like a swarm of mosquitoes, but if a few of the fish align themselves, the alignment spreads through the rest of the group. The best way to maintain a comfortable distance from your neighbors in an aligned group is to align yourself.

It's not clear what causes the first few fish to align, Couzin said, though they seem to do it in response to a predator.

Fish also can change positions within a school without giving the matter much thought. Near the center, there would be fewer encounters with either prey or predators. A fish near the center would be relatively safe but hungry, Couzin said, so hunger might cause the fish to space itself a little farther away from its neighbors. This would have the effect of moving it to the outside of the pack.

A fish that just ate could do the opposite to get back to the center.

Couzin's latest work, published this year, followed creatures with even less native intelligence than fish - ants. But in large groups, army ants can be very organized.

Found in the jungles of Panama and other tropical areas, army ants weave their bodies together to create a makeshift nest, Couzin said. A colony may have a million members, and every morning about 200,000 foot soldiers leave on a food run.

Couzin was able to follow the ants thanks to a special software package he developed to analyze the video he shot, helping to pick out the almost-imperceptible images of the ants and highlight them.

Practically blind, the ants rely on their sense of smell to stay aligned on long trails. They emit tiny amounts of a chemical called a pheromone, which the other ants recognize as a kind of lane marker. The ants follow the trail with their scent-sensitive antennae.

So finely tuned are they, that if an antenna on an ant's left side starts to pick up less scent than on the right, the ant veers to the right to get back to the center of the path. If it senses less pheromone on the right, the ant will angle to the left.

That leads to the precise lines that the ants take on their marches. When they kill their prey they turn around and go back along the same trail.

In principle, that could lead to collisions with outgoing ants, but it doesn't, Couzin said. Ants coming back turn at a sharper angle than outgoing ones, so when they get close to a head-on collision, the returning ants move inward, forming lanes inside the original narrow line.

In Couzin's video, it all looks very polite and organized.

Couzin sees other examples of self-organizing all over the animal kingdom. Soon, we may see it in robots as well.

Last month, Princeton's engineering school invited Couzin to discuss his work in ants and previous studies of schooling fish. The engineers are working to build a school of fishlike robots called "gliders" that would work together to gather data on the oceans.

Others are interested in land-roving herds of robots that might explore Mars or sniff for chemical weapons in the desert. The advantage to using many small robots, the engineers say, is that they can lose one or two individuals and still accomplish a given mission.

The Princeton engineers and their colleagues plan to launch their first trial run this summer, off the California coast. The robots will record data on density of plankton, water chemistry and currents. Understanding how real fish and other living things organize themselves could help them figure out how best to program the robots, said Naomi Leonard, a professor of mechanical and aerospace engineering. "We're trying to do things like real fish do."