Targeted memory reactivation in human REM sleep elicits recurrent, temporally compressed reactivation

Memories are reactivated during non-rapid eye movement (NREM) sleep, but the question of whether equivalent reactivation also occurs in rapid eye movement (REM) sleep is hotly debated. To examine this, we used a technique called targeted memory reactivation (TMR) in which sounds are paired with learned material in wake, and then re-presented in subsequent sleep to trigger reactivation. We then used machine learning classifiers to identify TMR-induced reactivation in REM. The reactivation we measured was temporally compressed by approximately five times during REM compared to wakeful performance of the task, and often occurred twice after a single TMR cue. Reactivation strength positively predicted overnight performance improvement and was only apparent in trials with high theta activity. These findings provide strong evidence for memory reactivation in human REM sleep after TMR as well as an initial characterisation of this reactivation.