Sleeper's Wake

sensory information and to modify existing associations mune function, endocrine function, and more. Long-term deprivation, whether induced experimentally in animals with appropriate deliberation. or induced by disease in humans, invariably leads to Naturally, it is tempting to speculate about the mecha-death. So our relative ignorance regarding the function nistic role of NCS-1 in plasticity. However, the NCS fam-of sleep, especially after years of studying its rhythms, its ily of proteins is sizeable, diverse, and largely unex-electrical signatures in the brain, its various stages, and plored. C. elegans appears to have five NCS-related its biochemistry, is all the more striking. genes and humans have perhaps a dozen (Burgoyne One of the proposed functions of sleep for which there and Weiss, 2001), with different members implicated is a considerable amount of evidence is memory consoli-in processes as diverse as guanylyl-cyclase regulation dation. According to this view, sleep (and in particular (Palczewski et al., 1994), K ϩ channel modulation (An et REM sleep) is a period of little sensory input during al., 2000), and protein kinase inhibition (Chen et al., which the brain rehearses or replays events or newly 1995). A role in control of synaptic strength (e.g., Pongs learned procedural tasks. The replay would then be re-et al., 1993), though mechanistically not yet understood, quired for these memories to be solidified in the brain. is the most promising in explaining the current results. This idea is supported by a number of studies in both This NCS-1 function is likely to be in the interneuron humans and animals showing that disrupting REM sleep AIY, but whether it functions pre-or postsynaptically, reduces subsequent performance on simple memory at what synapse, and by what biochemical mechanism tasks (Karni et al., 1994; Stickgold et al., 2000). Evidence are all unknown. The pump has been primed, but much of the replay itself comes from an elegant paper by Louie remains to be investigated. and Wilson (2001) recently published in Neuron and widely reported in the popular press. These authors re-James H. Thomas corded from several place neurons in the hippocampus Department of Genetics as an animal ran a stereotyped path in a circular maze University of Washington to receive a food reward. Given the repetitive path the Seattle, Washington 98195 animal took, the place neurons also fired in a stereotyped sequence as the animal repeatedly entered the of memory consolidation, they are nevertheless sugges-(placing the animal on …