Scurrying Roaches Outwit Without Their Brains

San Diego,California—From 4 to 8 January, agile roaches, swimming robots, and digging reptiles demonstrated the synergy between robotics and biology.

To an urbanite, it's a depressingly familiar scene: You flip on the light, and there's a cockroach, zipping across the cluttered countertops, scooting along a wall, and disappearing behind the sofa. Cockroaches are master escape artists. Two studies combining math, engineering, neurobiology, and biomechanics have begun to tease out the secret of this pest's success: conservative

To think or not. On obstacle courses, cockroaches are on autopilot (upper). But this ante-naed robot (lower) has shown that running along walls takes brains.

use of brainpower. According to the new research, a cockroach taps the brain when sticking close to walls but skips nervous system control during excursions across countertops and other uneven surfaces. "It can go on its own without a lot of sensory input," says Robert Full of the University of California, Berkeley, who led the work.

In 2002, as part of their effort to understand cockroach locomotion over flat surfaces, Full and his colleagues tied miniature cannons onto cockroaches' backs. When they fired the cannon to knock the treadmill-running insects off balance, the researchers discovered that the bugs seemed to recover too fast for their muscles to be controlled by nerves (Science, 6 September 2002, p. 1643).

To follow up, Simon Sponberg, a graduate student in Full's lab, tracked the neuromuscu-lar activity of cockroaches as they scrambled through an insect-scale obstacle course. "We usually think about these complicated leg movements as being coordinated neuro-muscular interactions," says John Bertram, a biomechanicist at the University of Calgary in Canada.

But that proved not to be the case. Spon-berg's colleagues first used mathematical modeling to show that an insect relying on the natural springiness of its legs could run the obstacle course without peripheral nervous system guidance. Next, they modified the control program of a cockroach-inspired robot so that it ran without such guidance; it did fine on an obstacle course. Sponberg then monitored the electrical activity of cockroach leg muscles and the nerves working them as the insects sprinted across both flat and rough terrain. The pattern of electrical activity was the same on both terrains, indicating that no additional neural control is used to navigate complex environments, he reported at the meeting. The work "has revealed that the mechanical system [legs, etc.] is a complex, dynamic system with a mind of its own," says Devin Jindrich, a comparative physiologist at the University of California, Los Angeles.

Such independence simplifies locomotion, as the brain doesn't have to keep track of either the legs or the obstacles. If designed properly, robots too could conserve brainpower, adds Jindrich. "That allows you to free up control for other things that might be more difficult," he notes.

However, brainpower is crucial to running next to a wall, another typical behavior for cockroaches, says Noah Cowan, now an engineer at Johns Hopkins University in Baltimore, Maryland. Cowan recently blindfolded cockroaches, forcing them to use just their antennae for guidance. Using high-speed video of roaches running next to walls, he concluded that the insects monitor the bend of an antenna as it touches a wall. If the antenna bends back too much, the body is heading too close; when the antenna is straight, the insect is too far away. Sensing these differences, the brain signals muscles and adjusts the insect's orientation to the wall accordingly. When he gave an antenna-

laden robot that capability, however, it didn't stay close to the wall at all.

After more observations of the live specimens, Cowan realized that a cockroach also factors in its speed. It determines velocity based on how quickly the antenna bends and unbends, input that adds another degree of control for the behavior. With that added feature, the robot excelled as a wall runner. "A combination of these two control systems was absolutely necessary," says Bertram. If only such understanding of how roaches use—or don't use—their brains would make us smarter about catching them.

With Flippers,Two Can Equal Four

Researchers trying to model how a beast that vanished millions of years ago swam through oceans have discovered that more isn't always better when it comes to flippers.

Intuitively, two pair of fins—as in those used by the large, extinct reptiles called ple-siosaurs—would be faster than one pair. But then why do many modern aquatic animals usually use just two, with the other two limbs reduced in size or eliminated all together? Seals, for example, evolved from a four-legged terrestrial ancestor but now depend on just modified hind feet for locomotion. Similarly, sea lions use mainly their front flippers.

Using a robot, a team of engineers and biologists has begun to resolve not just how plesiosaurs swam but also the pros and cons of two versus four flippers. Their preliminary conclusion: Two limbs are good for a steady swimmer, and four are better for starts and stops, John Long, a vertebrate physiologist at Vassar College in Poughkeepsie, New York, reported at the meeting.

Long's colleagues Charles Pell, Brett Hob-son, and Matthew Kemp of Nekton Research LLC in Durham, North Carolina, designed and built their robot over the past 9 months, dubbing it "Madeleine." She looks a little like a turtle and can swim forward or backward. Each side has a front and back "flipper"—flexible flaps that can move in sync or independently. The robot can roll and wiggle side to side as well as tilt its body up and down. "By simply turning on various combinations [of the fins], they can get different kinds of locomotion,"

Hang 10. Researchers are using this surf-swimming robot to learn about underwater locomotion.

says Frank Fish, a functional morphologist at West Chester University in Pennsylvania.

A computer onboard Madeleine lets the researchers determine the efficiency of locomotion in a way not yet possible in living organisms, says Adam Summers, a comparative physiologist at the University of California, Irvine: "It's not a good mimic of a truly biological system, but it provides a platform to understanding the mechanics of multiple [limbs]."

Long has now looked at the possible mechanics of swimming plesiosaurs, which plied the oceans between 248 million and 65 million years ago. Researchers have had different ideas about how these creatures' four flippers worked. To begin to characterize a ple-siosaur's stroke, Long's team has varied the speed ofthe robot's flapping, the time between each stroke, and whether the fins worked independently or together as a twosome or foursome. "We asked the robot how might the [ple-siosaur's] four limbs interact," says Long. They also compared those interactions against the dynamics of using just two limbs.

When Madeleine was swimming steadily, two fins were as good as four. "Adding fins did not produce faster [motion]," Long explains. Water swirling offthe front fins interfered with the thrust provided by the second set, preventing any boost from the extra fins.

Four fins came in handy for starting and stopping, though. When the robot begins to swim, its front fins have not yet created any turbulence, enabling the back fins to work efficiently. And four fins were better at stopping than two, thanks to the added resistance created by the extra pair. Long suggests that ple-siosaurs, with their four limbs, "may have been good starters and stoppers" who ambushed prey rather than chasing them down.

Robert Full, an integrative biologist at the University of California, Berkeley, worries that Long, Pell, and their colleagues' approach is too simplistic to reveal how plesiosaurs swam or even that two fins are better than four. He suggests that they need to do more mathematical modeling and make more of an effort to incorporate biological data into the robot's design.

But even with the robot's shortcomings, Summers has high hopes for Madeleine. "I'm very impressed," he says. "I think there are many more interesting questions ahead for this robot."

More Than One Way to Dig a Tunnel

Digging with one's nose is no small feat. Snakes and other legless animals do just that, typically relying on tough skulls, long slender heads, and strong trunk muscles to pound their way through soil. But researchers have recently found surprising exceptions, including a snake that tunnels by "wiggling" its nose and another reptile that uses its head but not its body to push forward. "We've had a lot of traditional ideas about how limbless animals burrow," says James O'Reilly, a functional morphologist at the University of Miami, Florida. "They all have to be revised."

Most legless burrowers are small, with tiny eyes and a narrow, pointy snout fused to the skull to form a battering ram. The shield-nosed cobra (Aspidelaps scutatus) doesn't fit that mold. It's small, about 50 centimeters long, but its eyes are large, its head is broad, and its snout is just loosely connected to the rest of the skull. In addition, it has a large, flat scale—the shield—at the tip of its nose. None of this hints at the snake's talent: "It doesn't look like it would be able to burrow," says Adam Summers, a comparative physiologist at the University of California, Irvine.

It digs quite well, however, just not the way people would have thought, Alexandra Deufel, a functional morphologist at Minot University in North Dakota, reported in San Diego. She put the shield-nosed cobra into an aquarium filled with moist sand and video taped the snake s progress as it tunneled along the bottom.

The cobra's shield can move independently of the head, she found. To begin, the snake arches its neck, lays the shield flat on the ground, and moves its head from side to side, throwing the shield back and forth. The shield wiggles just slightly—about 10 degrees off center—but enough that it can shove a little dirt out of the way with each nod of the head. "That's a completely novel mechanism," says O'Reilly. Adds Summers: "It's awesome to see a snake that can wiggle its nose."

O'Reilly found a different digging strategy in a worm lizard, a reptile that is neither a snake nor a true lizard. Unlike most worm lizards, Leposternon microcephalum uses its head, with little help from its body, when traveling underground. "We were really surprised," O'Reilly says.

Normally an elusive study subject, hundreds of these worm lizards were recently captured by O'Reilly's Brazilian colleague Nelson Jorge da Silva of the Catholic University of Goias as the creatures tried to escape the rising waters behind newly constructed dams. That allowed O'Reilly to study their digging habits more closely. For example, he set up a test tank where the reptile pushed against a force sensor as it tunneled through the soil. The worm lizard pounded its head against the soil around it, presumably to pat the soil down and make a clear path for the rest of the body. In most other legless dirt dwellers, the body provides the momentum that drives the animal forward. Typically, muscles along the body contract, bracing it against the sides of the burrow and enabling the head to ram into the dirt ahead. But L. microcephalum doesn't really have those muscles, O'Reilly and his colleagues found.

Instead, as researchers discovered when the worm lizard was placed in a very short artificial tunnel, the work is done by using muscles in the head. In such a setting, most limbless animals lose traction because their body needs walls to brace against—but not the worm lizard, which braces its skull against the burrow and thrusts forward and upward. "One of the most surprising [findings] is that these guys could generate all this force with virtually none of their body," says Summers. Now that's using your head.

-Elizabeth Pennisi

Supplements For Diabetics

Supplements For Diabetics

All you need is a proper diet of fresh fruits and vegetables and get plenty of exercise and you'll be fine. Ever heard those words from your doctor? If that's all heshe recommends then you're missing out an important ingredient for health that he's not telling you. Fact is that you can adhere to the strictest diet, watch everything you eat and get the exercise of amarathon runner and still come down with diabetic complications. Diet, exercise and standard drug treatments simply aren't enough to help keep your diabetes under control.

Get My Free Ebook

Post a comment