Creative problem-solving after experimentally provoking dreams of unsolved puzzles during REM sleep

Dreams have arguably been a source of creative insight for millennia. The specific assertion that dreams during rapid eye movement (REM) sleep promote creative problem-solving, however, has only anecdotal support, lacking strong empirical support from rigorous studies. Experimental manipulations of dream content have been confounded by waking components, such that any boost in creative problem-solving could be attributable to waking cognition rather than sleep cognition. Likewise, correlational evidence cannot unequivocally establish that dreams cause insights. Evidence that memory reactivation during sleep promotes creative problem-solving is also insufficient for implicating dreaming <i>per se</i>. Better methods for directly manipulating REM-sleep dreaming are needed. Here, we studied individuals who frequently have lucid dreams-realizing they are dreaming while still asleep. Participants slept after failing to solve several puzzles that had unique soundtracks, and they were instructed to continue working on a puzzle if they heard its soundtrack in a dream. Half of the soundtracks were played during REM sleep to reactivate memories of corresponding puzzles, with the goal of biasing dreams to connect with those specific puzzles <i>versus</i> the remaining puzzles. Those sound cues reliably increased dreaming about the associated puzzles. Furthermore, a post-hoc analysis showed that, for participants with an increase in cue-related dreaming, cues boosted later puzzle-solving. We thus expanded on a well-known phenomenon, that sounds can be incorporated into dreams and can change dream content, by substantiating experimental procedures to align dreams with the search for creative answers to specific challenges. Results highlight that REM dreams can contribute to next-day problem solving.