Evidence That Dreams Help Us Learn

Dreaming about something helps you to learn it more thoroughly.

Posted Nov 25, 2018

One of the most interesting lines of research in dream psychology these days are the so-called “dream incorporation” studies. Incorporation of novel learned material into dream content significantly predicts success at permanent retention or learning that novel content (Fiss et al., 1977; De Konnick et al., 1990; Stickgold et al., 2000; Fosse, Fosse, Hobson, & Stickgold, 2003; Wamsley, Perry, et al., 2010; Wamsley, Tucker, et al., 2010; Wamsley et al., 2012; Schoch et al., 2018; Wamsley and Stickgold, 2018). If you dream about something you studied that day you are more likely to retain that information the next day. Improved performance on learning novel materials is significantly associated with the extent of dream incorporation—the greater the number of elements of the learned materials incorporated into dreams, the stronger the acquisition of those novel materials in subsequent daytime performance tests. For example, De Koninck et al., 1990 found that second language acquisition scores were predicted by a higher frequency of dreams that incorporated elements (words, phrases etc) of the second language into dream content. Wamsley et al. (2012) demonstrated that learning to master a virtual maze navigation task was significantly associated with dream incorporation of elements of the maze. Greater incorporation predicted greater mastery. Learning-related dream incorporation into REM dreams is strongest on the night after and from 5–7 nights after learning, reflecting “day residue” and “dream-lag” effects, respectively (Nielsen et al., 2004; van Rijn et al., 2015).

The neurobiologic basis of these dream incorporation effects likely involves sleep-dependent synaptic plasticity and memory consolidation processes. A large number of studies have now demonstrated differential performance improvements (e.g., from training to retest) on a large variety of tasks over an intervening period of sleep vs control periods of wakefulness. Human neuroimaging studies incorporating sleep-dependent memory paradigms suggest that after new learning, there is physiological reactivation of brain areas recruited during learning (Peigneux et al., 2003, 2006; Oudiette and Paller, 2013; Oudiette et al., 2013; Fogel et al., 2017). Single cell recordings in animal studies have shown that during sleep there is a physiological replaying of the neural representation formed during waking learning (e.g. exploration of a maze). There is some evidence that this replay of neural representations of learned materials occurs in humans as well and is reflected in the content of our dreams (Stickgold et al., 2000; Wamsley et al., 2010; Kusse et al., 2012; Wamsley, 2014; 2018). “Targeted memory reactivation” refers to the use of cueing techniques to facilitate reactivation and replay of neural representations or memories during sleep to promote later waking recall of those memories (e.g., Cellini and Capuozzon, 2018, for review). For example, playing selected words or sounds during sleep that were previously paired with to-be-learned words during a learning phase before sleep results in better recall of the word pairs post-sleep. Presentation of cues during sleep that had been previously associated with a stimulus during wake results in better recall of that stimulus during subsequent waking recall tests. These techniques have been used to improve memory performance in healthy people and in patient populations. By re-exposing sleeping subjects to odors, words, or tones (i.e., cues) associated with newly learned neutral or emotional memories, emotional memories were reactivated and more readily recalled later. It should be possible to use targeted reactivation techniques to reinforce dream images of previously studied material. Then we would be harnessing the power of dreams in a whole new way.