How a sound played during sleep can strengthen a daytime exercise
Imagery rehearsal therapy asks a patient to take a recurring nightmare and deliberately rewrite its storyline into something more positive, then rehearse that new version. The idea is to build a competing, less frightening memory that can take the place of the old one. Targeted memory reactivation tries to give that new memory a boost at the moment the brain is most receptive: while the person sleeps.
The procedure is simple in principle. During the therapy session, while the patient is generating the positive ending to their nightmare, a particular sound is played. That sound becomes a tag attached to the new memory. Later, during REM sleep, the same sound is played again through a wireless headband that automatically detects sleep stages, so the tagged memory gets reactivated without waking the sleeper [1]. In the trial, half the patients went through this sound pairing and half did not, while everyone did the therapy every evening at home for two weeks [1].
The comparison matters because it isolates the sleep cue. Both groups received the same therapy and the same amount of practice; the only difference was whether a sound had been linked to the rewritten dream and replayed during REM sleep. The group with the cue ended up with less frequent nightmares and more positive dream emotions after two weeks [1][2]. That pattern is consistent with the cue strengthening the therapy-related memory rather than the therapy simply working on its own.
What actually changed: fewer nightmares, warmer dream emotions
The main outcome was how many nightmares patients had per week, measured after two weeks of therapy [1]. Patients whose therapy was paired with the sleep cue had fewer of them than patients who received therapy alone [1][2]. The benefit did not vanish when the study ended: nightmares were still reduced at a three-month follow-up [1].
The second change was emotional rather than numerical. Patients in the cued group reported more positive emotions in their dreams [1][2]. This is a distinct result from counting nightmares. A person could have fewer bad dreams without their dreams becoming any more pleasant; here the dreams appear to have shifted in tone as well as in frequency. The authors frame this as evidence that reactivation during REM sleep can modulate emotion in dreams, not merely suppress the bad ones [1].
A separate line of work offers a possible way to describe that shift. Using a language model to categorize dream diaries, researchers identified what they call cathartic dreams: dreams with a dynamic plot and emotional twists in which negative feelings are expressed and then fade, which they suggest may serve an emotion-regulation function [3]. They observed an increase in this kind of dream during two weeks of imagery rehearsal therapy combined with targeted memory reactivation during REM sleep, and the increase correlated with a drop in depression scores in patients with nightmares [3]. This is a proposed functional interpretation, not a demonstrated purpose, and it describes a different feature of dreams than the nightmare counts and positive-emotion ratings in the treatment trial.
Who this has been tested in, and how far the claim reaches
The strongest direct evidence comes from a small group: 36 patients diagnosed with nightmare disorder, randomized into cued and uncued therapy [1]. Small samples can produce results that later studies do not fully reproduce, so the size of the group is worth keeping in view alongside the size of the effect. The follow-up at three months helps, because it shows the reduction was not a one-night fluctuation [1].
The population also matters. These were people with nightmare disorder, not general dreamers hoping for nicer dreams, and the intervention was delivered as a clinical treatment alongside nightly homework. Nothing in this evidence indicates that playing a sound during sleep would change dream content in someone who has not first done the rehearsal that gives the sound something to reactivate.
Related work extends the question in two directions without settling it. Imagery rehearsal has also been combined with art therapy for post-traumatic nightmares, where making images rather than talking through the trauma is proposed as a way to reach memories stored in non-verbal form; that review describes the approach as promising but calls for further research [4]. And a study of imagery rehearsal therapy plus targeted memory reactivation in post-traumatic stress disorder examines clinical and neurophysiological effects, but its abstract does not report nightmare frequency or positive dream content as outcomes [5]. So whether the sleep-cue boost carries over to trauma-related nightmares, and how it relates to broader PTSD symptoms, remains an open question rather than a demonstrated extension.
The practical answer for someone weighing this
If the question is whether the combination works better than the therapy alone, the trial points to yes on both counts the question asks about: fewer nightmares and more positive dream emotions, with the nightmare reduction holding at three months [1][2]. The mechanism is plausible and specific, since the sound is tied to the rewritten dream and delivered during REM sleep, the stage in which nightmares themselves occur [1].
What the evidence does not yet provide is a large, independent confirmation, or a clear picture of who benefits most. The cathartic-dream findings suggest one possible route by which dream emotions shift, and the correlation with lower depression scores is intriguing, but it comes from a proposed framework rather than a controlled test of that mechanism [3]. For a patient considering this, the reasonable reading is that the sleep cue appears to add something to imagery rehearsal in the short term, while the longer-term and broader clinical picture is still being filled in.
Sources used in this answer
Enhancing imagery rehearsal therapy for nightmares with targeted memory reactivation
In 36 patients with nightmare disorder, pairing a sound with the positive rewritten nightmare during imagery rehearsal and replaying it during REM sleep led to fewer nightmares and more positive dream emotions than therapy alone after two weeks, with a sustained decrease at three months.
Combined treatment of nightmares with targeted memory reactivation and imagery rehearsal therapy: a randomized controlled trial
The combined targeted memory reactivation and imagery rehearsal treatment reduced nightmares and increased positive dream emotions relative to imagery rehearsal alone.
The cathartic dream: Using a large language model to study a new type of functional dream in healthy and clinical populations
Cathartic dreams, identified with a large language model, increased during two weeks of imagery rehearsal therapy plus targeted memory reactivation during REM sleep, and that increase correlated with decreased depression scores in patients with nightmares.
Imagery Rehearsal Based Art Therapy: Treatment of Post-traumatic Nightmares in Art Therapy
A framework combining imagery rehearsal with art therapy is described as a promising method for post-traumatic nightmares, proposed as especially useful for patients who benefit least from verbal exposure techniques.
Clinical and neurophysiological effects of imagery rehearsal therapy and targeted memory reactivation in post-traumatic stress disorder
A study examined clinical and neurophysiological effects of imagery rehearsal therapy and targeted memory reactivation in post-traumatic stress disorder, but its abstract does not report nightmare or positive dream content outcomes.