Here's one more thing to worry about when you're lying awake at 4:00 a.m.:
If you don't get back to sleep, you risk forgetting whatever it was you learned yesterday, impairing your ability to learn new things tomorrow, and preventing yourself from extracting general concepts from a set of examples. Oh, and if you're a gambler, you're more likely to make reckless bets.
How's that for a nice soothing cup of hot cocoa?
The Society for Neuroscience is holding its annual meeting this week in Washington, D.C. The city is lousy with neuroscientists, more than 30,000 of them. (If you're in D.C. right now, a word of advice: avoid Chinatown and anywhere else near the convention center until at least Thursday.)
Sleep is one of the hot topics in neuroscience these days. Robert Stickgold of Harvard, who has been studying sleep and memory for decades, summed it up yesterday morning: "The last ten years have been phenomenal in our growth of understanding of the multiple functions of sleep." What's it good for? "Enhancing, consolidating, and improving memories and our understanding of what those memories mean to us," he said.
So, for instance, Dennis McGinty of UCLA presented evidence that, in rats, sleep fragmentation—the sort of fits-and-starts sleeping that is one consequence of sleep apnea—prevents rats from learning how to escape from a maze.
Chiara Cirelli of the University of Wisconsin-Madison, presented an interesting hypothesis about the benefits of sleep: it lets our synapses relax. Synapses are connections between neurons that strengthen or weaken with experience. Strong synapses are energetically expensive (maintaining our brains takes a ridiculous amount of energy), and sleep allows synapses to weaken. The net result? The next day we're ready to learn again. It's an intriguing idea, and her early results with rats and mice have been well received.
Moving to people, William Fishbein of CUNY showed that naps are good for "relational memory," which makes more sense once you hear what he and his students did. They showed a bunch of Chinese characters and their definitions to non-Chinese speakers. On the list were "mother," "maid," and other words that referred to females, as well as distractor words that weren't ultimately part of the test.
Then they let their subjects take a 90-minute nap, or not.
Afterwards, the researchers showed the subjects the character for "princess" and asked them to choose what it meant from a multiple-choice list. People who had taken a nap were more likely to figure out the pattern: the left-most segment of those characters means "female."
Another study showed that people who gamble all night long are in trouble. Vinod Venkatraman, a grad student at Duke, found that sleep-deprived people were more likely to be "gain seeking" in a gambling task—that is, they made stupid bets overall, driven by a desire for a big payoff.
Sometimes brain research is subtle or contradictory or confusing, but not so much when it comes to sleep's effects on learning and memory. Overall the results in this field (and this week's results are just a sampling) are robust, reliable and pretty overwhelming.