Sleep is a vital part of our existence–we spend a third of our lives asleep, our memory and motor skills fade with lack of sleep, various diseases are associated with disorders of sleep–and yet no one truly knows why we need it. Scientists at the University of California, San Diego, hope to change that. They have developed a method of analyzing brain activity during sleep, which they say is much simpler than current techniques and could potentially be used to study a myriad of sleep disturbances.
“Sleep is a major health problem and scientific mystery,” says Terry Sejnowski, a head of the Computational Neurobiology Laboratory at the Salk Institute for Biological Studies, in La Jolla, CA. “We have a tool here that might help us get to the bottom of that mystery.”
To be diagnosed with a sleep disorder, a patient typically spends the night at a specialized clinic, hooked up to machines that record brain activity, muscle activity, and other factors. A technician or doctor then pores over the data, searching for signs that the patient’s peaceful slumber has gone awry. However, results of the test can vary widely, depending on who analyzes the data.
A new analysis method developed by Sejnowski and Philip Low, a graduate student at the Salk, could make the process much easier and more automated. The researchers have created an algorithm that can detect subtle but statistically significant changes in brain activity. The result is a program that can differentiate the phases of sleep, such as REM (rapid eye movement) sleep, which is typically when we dream, and deep sleep, using less data than other methods do. “The technique has a unique temporal resolution that may be able to define sleep states more accurately,” says Jean-Paul Spire, a neurologist and scientist at the University of Chicago who was not involved in the research.