Dream a little dream of ... huh?

Even if you can't recall a recent event, might you still dream about it? According to a new study, the answer -- a surprising yes -- is raising a slew of questions about where dreams come from, what they mean and what role they play in learning and memory.

Dr. Robert Stickgold, a psychiatrist at Harvard Medical School in Boston and the lead author of the research effort, says the combination of the popular computer game Tetris and a group of volunteers, including some with amnesia, has paved the way for future scientific studies of dreams.


"The studies with the amnesiacs have given us for the first time the ability to look at where in the brain our dreams come from," Stickgold says. "It's sort of like a 'Where's Waldo?' experiment."

In the study, reported this month in the journal Science, Stickgold and his team spent three days training 27 volunteers to play the computer game Tetris, which requires spatial reasoning to properly arrange a series of falling blocks. The study included 10 Tetris experts, 12 novices with no previous experience and five people with amnesia who had to be retaught the game each time they played it.


After morning and evening playing sessions on the first two days, 17 of the volunteers reported dreaming of falling or rotating Tetris blocks in the first hour of their slumber. To the surprise of the scientists, those dreamers included three volunteers with amnesia who didn't remember playing the game or even the researcher who taught them.

Since such sleep-onset dreams commonly contain isolated snippets of recent experiences, researchers had guessed that they originate from memories stored in the hippocampus, a region of the brain. But people with amnesia brought on by hippocampal damage are unable to remember recent events. They can, however, remember isolated facts stored as memories in the brain's outer neocortex.

Stickgold says the study results suggest that memories of facts, such as the shapes and sizes of the Tetris blocks, were the true source of the sleep-onset visions dancing in the game players' heads. And he predicts the finding will also hold true for the truly weird dreams we encounter late at night.

"Part of the reason our dreams are so bizarre and disconnected is that we're not filtering that through our episodic memories that hold everything together," he says.

Upon waking from an afternoon nap, for example, one of the volunteers reported dreaming about Tetris pieces falling onto a garden path. The seemingly disjointed nature of the image, Stickgold said, is a hallmark of a late stage of sleep characterized by rapid eye movement and intense dreams.

The dream can be explained, he said, by the brain's attempt to link two related topics, such as ordered game blocks and the ordered nature of landscape architecture.

"It's like the brain is saying, 'Wait a minute, guys, I know this seems bizarre and illogical, but it kind of fits together,' " Stickgold said.

Figuring out what to do with incoming data is "probably the most difficult task that the brain has to perform," he said, a learning process he says is a critical function of sleep -- and dreams.


Among other findings in the study, two of the expert Tetris players reported dreaming about earlier color versions of the game they had played instead of the black-and-white version used in the experiment. Stickgold said their mental association of related images supports his notion that dreams act as learning aids.

Richard Haier, a psychologist in the department of pediatrics at the University of California, Irvine, called Stickgold's study "intriguing," although he noted the small study size warrants caution when interpreting the results.

For Stickgold, the popularity of power naps and longer siestas is telling. "Sleeping in general, and dreaming in particular, is giving the brain the chance to disconnect from recent events and just look at the raw facts," he said. "It's what we mean when we say, 'I'm going to sleep on it.' "