The Effects of REM Sleep Deprivation on Spatial and Reversal Learning.

Abstract

Typically, when REM sleep restriction is applied during learning (concurrent learning) performance is impaired. It is unclear how REM sleep restriction can alter other forms of spatial learning (e.g. reversal learning). For my dissertation I studied the effect of concurrent REM sleep restriction on both initial spatial learning and reversal learning, as well as the effect of prior REM sleep restriction on subsequent reversal learning in the Morris water maze. When using 12 training trials per day, I found that REM sleep restriction concurrent with either initial spatial learning or reversal learning were not affected. However, prior REM sleep restriction resulted in performance deficits during subsequent reversal learning. When using 4 training trials per day, I found that again REM sleep restriction did not affect concurrent reversal learning. In contrast, REM sleep restriction resulted in the typically reported deficits during initial spatial learning. Additionally prior REM sleep restriction was associated with performance enhancements during subsequent reversal learning. My results suggest that concurrent reversal learning is protected from the detrimental effects of REM sleep restriction. Across my dissertation, I identify an interactive relationship between the number of training trials, or learning load, and REM sleep restriction to modulate performance. Though REM sleep restriction does appear to alter learning, the performance deficit may not be measurable during the initial learning experience if the training session is sufficient to produce near asymptotic performance, but only on subsequent learning events. These behavioral findings support the hypothesis that REM sleep facilitates both the consolidation of incomplete learning and the desaturation of neuronal networks for subsequent learning purposes. Lastly, my studies emphasize the inability to ascertain the role of REM sleep when generalizing across learning, even when limiting the focus to spatial learning.