Flies sleep, despite what that buzzing in your ear might tell you. But sleep's purpose for flies--and why animals ranging from invertebrates to mammals such as humans do it--remains a mystery, particularly because simple rest alone seems to deliver similar restorative benefits to tired muscles. Recent studies have hinted that sleep might play a role in memory formation, and new research on fruit flies confirms this purpose and reveals some of the genetic mechanisms behind it.

Indrani Ganguly-Fitzgerald of the Neurosciences Institute in San Diego and her colleagues chose fruit flies for two reasons: they are known to sleep and the size and richness of their tiny brains depend on youthful experience. Flies that are isolated during the early days of their lives develop smaller brains with fewer synapses compared with their more sociable peers. And young flies require much more sleep than adults.

To test the link between these two facts, the neuroscientists isolated some young flies. Five days later, these loners slept significantly less than their socialized brethren, catching 15-minute naps during the day compared with roughly hour-long blocks of quiescence for the others. This sleep preference persisted over time. In the socialized flies, keeping them awake did not reset their sleep preference; they returned to their normal, longer sleep schedule once left unmolested. And the amount of sleep they needed rose in direct proportion to the number of other adult flies to which they were exposed.

The researchers also wanted to uncover the cause of this phenomenon, which could not be linked to either the amount of space of the flies home or sexual activity. Nor could it be attributed to general activity; the motion of both the isolated and socialized flies was equal. But when the researchers raised the flies in darkness or deprived them of the sense of smell, even socialized flies stopped sleeping as much. And when socially impoverished flies joined a group as adults, they began sleeping for longer periods; switching between the two social states consistently prompted changes in sleep.

Analyzing the brains of these flies, the neuroscientists found that isolated flies had one third less of the neurotransmitter dopamine, and artificially manipulating those levels could produce the desired sleep state. Further, by knocking out certain genes associated with memory, the researchers could change a fly's sleep requirements. They also found that trained flies--males who had been spurned in the past failed to court new females because of this memory--slept considerably longer than their untrained peers, and flies in training that were not permitted to sleep did not form the expected rejection memory. "Sleep and neuronal activity may be inexorably intertwined," the researchers write in the September 22 issue of Science.. And because humans and flies share much of their genomes, studying the insect's sleep and memory may allow us to finally understand our own. As Ganguly-Fitzgerald says: "Of all genes known to cause human disease, more than 60 percent are found in the fruit fly."