FIGURE 4.

The decrease in REM sleep with age parallels the decrease in plasticity. This result is directionally consistent with the present hypothesis that REM sleep is less necessary when neural circuitry is less flexible. (A) Across animal species, the fraction of sleep time spent in REM stage decreases (data in humans adapted from Roffwarg et al., 1966). (B) As just one example (see text for others), neuroplasticity diminishes with age in the human motor cortex, here measured with a paired associative stimulation protocol. Data from Müller-Dahlhaus et al. (2008). PAS_N20+2 stands for paired associative stimulation, tested at an interstimulus interval 2 ms longer than the N20 peak of the median nerve somatosensory-evoked potential 20 ms into the EEG event-related potential.