Nov. 20, 2008 -- If you've just learned a complex skill, here's how to do it better: Get a good night's sleep.
Practice may make perfect, but not if you haven't slept, suggest studies by Howard Nusbaum, PhD, David Margoliash, PhD, and colleagues at the University of Chicago.
Nusbaum, Margoliash, and colleagues studied the issue by teaching college students to play a first-person shooter video game called Unreal Tournament 2003. Only students who had never played more than 10 such games were allowed to participate. Most of the students in the study were female, 163 out of 207 participants.
First-person shooter games are complex. They require learning to use both hands to manipulate a computer keyboard and mouse to move through a virtual world in which they have to kill -- and avoid being killed by -- a variety of "bots." Players have to adjust to rapidly changing situations.
The basic study design was to see how well the students could play a second first-person shooter game -- Quake 3 -- after learning to play Unreal Tournament.
After being tested on their initial Quake 3 skills and then being trained to play Unreal Tournament, the students were divided into groups:
- The "AM control" and "PM control" groups received training in the morning or evening and immediate testing on Quake 3.
- The "12-hour wake" group received training in the morning but were tested on Quake 3 that evening, 12 hours later, without any sleep in the interim.
- The "12-hour sleep" group received training in the evening but were tested the next morning, 12 hours later, after getting a night's sleep.
- The "24-hour AM" and "24-hour PM" groups received training in the morning or in the evening and returned for testing 24 hours later.
Predictably, the "control" students played Quake better after learning the skills needed to play Unreal Tournament, regardless of whether they learned to play in the evening or in the morning. Time of day had no effect on learning this complex task.
The "12-hour wake" students tested 12 hours after training -- with no sleep -- were better than they'd been before training, but improved only about half as much as those tested immediately after training. It seems that their learning deteriorated over the course of the day.
But the "12-hour sleep" and "24-hour" groups -- which got a night's sleep before testing -- did just as well as those tested immediately after learning the skills.
"Sleep consolidated learning by restoring what was lost over the course of a day following training and by protecting what was learned against subsequent loss," Nusbaum says in a news release. "Sleep has an important role … in stabilizing and protecting memory."
The researchers suggest that during sleep, memory traces are replayed and modified while the brain is "offline." During this process, there is coordination of the sensory and motor systems involved in complex tasks learned during waking hours.
They report their findings in the November issue of the journal Learning & Memory.