Cover Story

Newborn babies are not very good at controlling their bodies. Like puppeteers testing which strings move which limbs, infants must find which neurons connect to which muscle fibers. The precise way babies do this remains a mystery, but new research suggests that much of the learning process may occur as infants slumber, says Mark Blumberg, PhD, a psychology professor at the University of Iowa.

"Quiet time spent asleep may provide a key opportunity for mapping relationships between muscle, spinal cord and brain," says Blumberg.

When awake, an infant's brain is constantly deluged with sensory stimulation and signals from muscles about their positioning, known as proprioceptive information, Blumberg notes. An intentional arm movement, for instance, requires the simultaneous firing of hundreds of muscle fibers, resulting in a glut of information returned to the brain. An infant who is just learning to make such a movement may need the opportunity to tug on just one string at a time, linking each individual muscle fiber to its single controlling nerve cell, and tracing the pathway back to the brain. The relatively inactive state of sleep may provide such an opportunity, he says.

In fact, the muscle twitches that are characteristics of active, or rapid eye movement (REM) sleep, may result from just such a system test. Babies trace new connections with twitches, while adults may primarily use them to maintain existing connections, researchers hypothesize.

Supporting this idea, Blumberg's lab has found that newborn rats spend about 70 percent of their time unconscious and twitching. What's more, sleeping infant rats twitch more than adult rats, perhaps because infant rats' bodies are changing and growing at an impressive pace, requiring the development of many new brain-muscle connections, according to fifth-year Iowa psychology graduate student Adele Seelke, who studies infant sleep with Blumberg.

"We know sleep is important, but we don't really know why it is important or how it works," says Seelke. "What this research is doing is giving us a clearer picture of this behavior that is so ubiquitous that it occurs in every species studied so far."

Two other teams of scientists are also finding support for the idea that twitches may help the nervous system organize itself. Researchers in Sweden have found that young rats develop reflexes through twitching. Another group, based in France, discovered that the brain responds to twitches with spikes of brain activity that may signify the recording of new information about the animal's developing body.

All three teams of researchers are converging on the same idea: A key purpose of sleep might be to twitch.

Twitch university

In addition to muscle twitches and rapid-eye movements, intense brain activity and dreaming occur during active sleep. This has led some to theorize that a sleeping person's eyes are tracking dream visions or that a napping dog's leg movements result from chasing imaginary rabbits. However, early research by Blumberg, published in a 1994 issue of Behavioral Neuroscience (Vol. 108, No. 6, pages 1,196-1,202) discounted those ideas by showing that many of the signals that cause muscle twitches originate from the spinal cord, not the dreaming brain.

What's more, Seelke recently found that the eye movements of active sleep result from the same sort of single-fiber muscle twitches as those that cause limb movements during sleep, according to a paper she published in the European Journal of Neuroscience (Vol. 22, No. 1, pages 911-920). These erratic eye movements do not look like the coordination motion of a wakeful animal's eye, she says.

Though they are probably not a byproduct of dreaming, muscle twitches in the body and in the eye may serve an important developmental function for infants, Seelke says. Certainly the amount of time infants spend doing it suggests some purpose. Human newborns, for instance, sleep around 75 percent of the time, and they tend to spasm and squirm as they slumber. The same goes for newborn rats, she notes.

"In my experience, younger rats, including 3-day-olds, twitch more often and with more intensity than 8-day-old rats, which in turn twitch more often and with more intensity than 15-day-old rats," says Seelke.

All that twitching may serve to help sleeping infants train their reflexes, says Per Petersson, PhD, a neuroscience lecturer at the University of Lund in Sweden. In a paper published in a 2003 issue of Nature (Vol. 424, No. 6,944, pages 72-75). Petersson and his colleagues found they could train infant rats' tails to move toward-rather than away from-a heat source by manipulating what the rat tail "learned" during sleep twitches.

The researchers lightly restrained 84 12- to 17-day-old rats, and let them fall asleep. As the rats slept, their tails twitched to the left and right. In a natural setting, a leftward twitch often results in the left side of the tail touching something, such as a littermate. For half of the rats, the researchers provided feedback opposite of what occurs naturally: They puffed air at the left side of the tail when it twitched to the right, and they puffed air toward the right side of the tail when the tail moved leftward. The other half of the rats received air-puffs on the correct side of the tail.

After just two hours of training, the rats that received the wrong feedback began to show a strange reflex-when a hot laser touched the right side of their tails, they would often move their tails toward the heat source, rather than away from it. The finding suggests that the reflex mechanisms in the spinal cord records sensory information learned through twitches, says Petersson. Such a system may also be at work at longer-range movement control, from the muscles to the brain, he says.

"I would like to think that the spontaneous movements and the self-testing procedure we found for the [rats] is a general mechanism whereby the central nervous system learns about the body it is in, no matter if it is in an elephant or in a mouse," says Petersson.

Mapping the body

Lending support to the idea that twitches may inform a newborn brain about its own body are findings from a team of researchers led by Rustem Khazipov, PhD, of the Mediterranean Institute of Neurobiology in France. They recently found that the spontaneous movements of newborn rats are immediately followed by prolonged bursts of brain activity known as spindles. In fact, reacting to twitches appears to be the primary brain activity of the animals, reported the team in a December 2004 issue of Nature (Vol. 432, No. 7,018, pages 758-761).

"There is only one single brain pattern right after birth of the rat, and it is similar to patterns you can record from a premature baby," says paper co-author Gyorgy Buzsaki, MD, PhD, a behavioral neuroscience professor at Rutgers University in New Jersey. "It is silence followed by spindle activity."

The fact that brain activity follows the muscle twitches suggests that the brain is recording information learned from twitching, such as the weight of the attached limb and its distance from other parts of the body. Also, the team found that the brain activity occurred in the somatosensory cortex, an area that eventually develops a map of the body.

While infants must work to map out their entire bodies, in adults the work is primarily finished. However, research on infant sleep may hold clues as to the purpose of sleep in adults, says Blumberg. If adult twitches are like those of infants, adult brains may be checking motor pathways and making necessary adjustments in their mental maps-perhaps if part of the body has been injured, says Blumberg.

However, such research is in its infancy, notes Petersson.

"There are only a few groups in the world working with the importance of spontaneous movements," he says. "But my impression is that the field is steadily growing. And it is exciting that sleep researchers, developmental neurobiologists and neurophysiologists working with sensory-motor systems are now focusing on this same question."

Further Reading

  • Blumberg, M.S., & Lucas, D.E. (1994). Dual mechanisms of twitching during sleep in neonatal rats. Behavioral Neuroscience, 108, 1196-1202.

  • Khazipov R., Sirota A., Leinekugel X., Holmes G.L., Ben-Ari Y., & Buzsaki, G. (2004). Early motor activity drives spindle bursts in the developing somatosensory cortex. Nature, 432(7018), 758-761.

  • Petersson P., Waldenstrom A., Fahraeus C., & Schouenborg J. (2003). Spontaneous muscle twitches during sleep guide spinal self-organization. Nature, 424(6944), 72-75.

  • Seelke, A.H., Karlsson, K.A., Gall, A.J., & Blumberg, M.S. (2005). Extraocular muscle activity, rapid eye movements, and the development of active and quiet sleep. European Journal of Neuroscience, 22, 911-920.