Can targeted memory reactivation test whether specific cued content appears in dreams?

How does TMR actually test for cued content in dreams?

TMR works by first associating a neutral sensory cue—often a sound—with a learning experience. Later, while the person sleeps, that same cue is replayed. If the cue reactivates the associated memory, researchers can look for traces of that memory in dream reports collected after awakenings. This design lets them ask a specific question: did the cued material show up in the dream, rather than just any random dream content? [2][5]

The method has been used with different learning tasks. In one series of studies, participants learned a virtual-reality flying task and task-associated tones were replayed during sleep; researchers then checked whether dreams contained task-related imagery such as flying or accelerating [1][3]. Another study paired pseudowords with picture categories and planned to cue one category during NREM sleep and another during REM sleep, with dream reports collected after awakenings [2]. These designs show that TMR can be set up to test for specific cued content, not just general dream changes.

When does cued content show up—and does sleep stage matter?

Timing is a central finding. In the VR flying studies, TMR did not have an immediate effect on dream content during the nap itself [1][3]. Instead, the effect appeared later: participants dreamed more about the task one to two days afterward when the cue had been played in REM sleep, and five to six days afterward when it had been played in slow-wave sleep, compared with participants who received no cueing [1]. That delayed pattern suggests that TMR can influence dream content, but the appearance of cued material may not be immediate.

Sleep stage also appears to shape how TMR affects dreams and performance. In one study, REM dreams that reactivated kinesthetic elements of the VR task—such as the feeling of flying or accelerating—were associated with greater improvement on the task than dreams that reactivated only visual elements or no task elements [3]. TMR itself did not change dream content in that study, but its performance benefits were greater when task-related dream reactivations were also present in REM sleep [3]. This suggests that the presence of cued content in dreams may be one part of a larger process, not a simple on-off switch.

Can TMR also change what dreams contain in clinical settings?

Beyond laboratory learning tasks, TMR has been tested as a way to strengthen therapy for nightmares. In a study of 36 patients with nightmare disorder, participants practiced imagery rehearsal therapy—rewriting a nightmare with a more positive outcome—while half were exposed to a sound paired with the positive imagery. During the next two weeks, that sound was replayed during REM sleep at home. The TMR group had fewer nightmares and more positive dream emotions after two weeks, and the decrease in nightmares was sustained after three months [4]. This shows that TMR can modulate dream emotions in a clinical context, which is indirect evidence that cued content can influence dreams.

However, the method has real limits. Dream reports depend on memory and awakening timing, and researchers have noted methodological inconsistencies across studies of dreaming and memory consolidation [5]. The 2025 study using pseudoword–picture pairs is still in progress—only 19 of 45 participants had completed the protocol at the time of its preregistration—so its confirmatory results are not yet available [2]. These caveats do not undermine the basic answer: TMR can be used to test for specific cued content in dreams, but the strength and timing of the effect vary with sleep stage and study design.

Sources used in this answer

Related research questions