The benefits of sleep seem obvious. And yet scientists have long debated precisely how it improves brain performance at the cellular level. One camp argues that sleep reduces the unimportant connections between neurons, preventing brain overload. Another camp maintains that sleep consolidates memories from the previous day.

A group of researchers recently tried to settle this debate by studying the larvae of a common see-through aquarium pet, the zebrafish. Like humans, zebrafish are active during the day and sleep at night. Unlike humans, zebrafish larvae are transparent, which allowed researchers to watch their brains as they slept. The researchers, led by Lior Appelbaum and Philippe Mourrain of Stanford University, tagged the larvae neurons with a dye so that active neuron connections, or synapses, appeared green, whereas inactive ones appeared black. Decreased synaptic activity would show that sleep pruned unnecessary memory connections, whereas memory consolidation would have a different pattern. After following the fluctuations of these synapses over the course of a day, the team found that the zebrafish did indeed have lower overall synapse activity during sleep. The researchers published their results in the journal Neuron, becoming the first to show the effects of sleep/wake cycles and time of day on the synapses of a living vertebrate. “Sleep is an active process that reduces the activity in the brain,” Mourrain says. “It allows the brain to recover from past experiences.” Without the synapse reduction that happens during sleep, he notes, the brain would not have the ability to continually take in and store new information.

But the debate is not yet settled. Among the team’s other findings was that not all neural circuits are affected by sleep in the same way. Learning and memory may benefit the most, Mourrain says. For this reason, the two hypotheses about sleep “may not be mutually exclusive,” says neuroscientist Jan Born of the University of Lübeck in Germany.  A resolution may not be far off; Mourrain and Applebaum’s new imaging technique will allow for more detailed study of the brain during sleep in years to come.

This article was originally published with the title Why Sleep Is Good for You.

4 Comments

1. 1. jtdwyer 03:30 AM 1/11/11

   I suspect that zebrafish do not acquire nearly as much new information during their waking hours as do humans. Perhaps zebrafish merely require reduced neural activity to replenish supplies of neurotransmitters, etc.
   
   If there is a separate physical facility within the human brain for the storage of new information, as implied by the concept of 'short term memory', there is most likely a process that migrates selected new memories out of this short term memory, integrating them into existing long term memory. Otherwise, short term memory would simply be reused daily as a 'temporary storage facility'.
   
   I understand that human short term and long term memory functions can be separately disabled by physical disruption to specific regions of the brain. This alone should indicate that some migration process is necessary. The performance of this activity during inactive periods of sleep would minimize its disruption to waking memory functions.
   
   These management concepts are commonly implemented in computer database and memory storage, likely intuitively modeled after natural process that occur in the human brain. This only a philosophical argument for an offline human memory optimization process, but I think it's a compelling one.

   Reply | Report Abuse | Link to this

2. 2. rgcorrgk 01:25 PM 1/11/11

   Very interesting work. “Sleep is an active process that reduces the activity in the brain,” Yes, this seems clear, after all when awake a great deal of brain activity deals with interactions with our surroundings; thus, of course, we would see a "selective" reduction, we would see "lower overall synapse activity during sleep". It would seem to me, however, that the reduced activity in sleep would be more of a nightly "default" setting, that is, "the synapse reduction" is a sort of forgetting process that makes brain room for future events (that is how I would state this). (Please bear with me I've been up all night - no joke.)
   Thus, for me, the likely way to deal with memories from the previous days brain work is to only keep the relevant ones! That is, most of adult prior awaking state memories are redundant, they are like most other days & need little or no sleep work to preserve them, as basically they are set in your head (the drive to work, for example). However, if a wrong way driver nearly killed you, on your way to work- 100% you will be doing some "consolidates" in your next sleep! And, if you wake up and recall it, it is a dream. (30 or 40 years ago I did research on dreaming, and had the good look of coming up with some elements). By the way, if you stay awake long enough you do fall in to sleep/dream states - and if you force yourself to stay awake the "consolidates" (learning) is/are impaired. My studies indicated to me that we ALWAYS dream ONLY about mental elements from our just prior awake state. Also, the elements seem to be what I call emotionally tagged. I would love to comment further, but, I've got to get some sleep! R. Carlson

   Reply | Report Abuse | Link to this

3. 3. rgcorrgk 01:31 PM 1/11/11

   Very interesting work. “Sleep is an active process that reduces the activity in the brain,” Yes, this seems clear, after all when awake a great deal of brain activity deals with interactions with our surroundings; thus, of course, we would see a "selective" reduction, we would see "lower overall synapse activity during sleep". It would seem to me, however, that the reduced activity in sleep would be more of a nightly "default" setting, that is, "the synapse reduction" is a sort of forgetting process that makes brain room for future events (that is how I would state this). (Please bear with me I've been up all night - no joke.)
   Thus, for me, the likely way to deal with memories from the previous days brain work is to only keep the relevant ones! That is, most of adult prior awaking state memories are redundant, they are like most other days & need little or no sleep work to preserve them, as basically they are set in your head (the drive to work, for example). However, if a wrong way driver nearly killed you, on your way to work- 100% you will be doing some "consolidates" in your next sleep! And, if you wake up and recall it, it is a dream. (30 or 40 years ago I did research on dreaming, and had the good look of coming up with some elements). By the way, if you stay awake long enough you do fall in to sleep/dream states - and if you force yourself to stay awake the "consolidates" (learning) is/are impaired. My studies indicated to me that we ALWAYS dream ONLY about mental elements from our just prior awake state. Also, the elements seem to be what I call emotionally tagged. I would love to comment further, but, I've got to get some sleep! R. Carlson

   Reply | Report Abuse | Link to this

4. 4. fcoaz 04:17 PM 1/11/11

   I found a lot of sense in your approach model to sleep brain activity.
   It also offers a compelling explanation to why psichopaths and Asperguer cases present a very few sleep hours pattern, three to four sleepping hours is very common for them.
   The issue there is their lack of emotions linked to congenital atrophy at some brain structures like the pituitary gland and other prefrontal areas.
   Not having to resolve many emotionally tagged memory elements from the awaked period they don´t need as much sleep time as normal emotional people.

   Reply | Report Abuse | Link to this