Reactivating a Relaxation Exercise During Sleep to Influence Cortical Hyperarousal in Individuals With Frequent Nightmares
Clara Sayk, A. Probst, F. Lange, S. Eickemeier, J. Amores, Hong‐Viet V. Ngo, M. Ehsanifard, Klaus Junghanns, Ines Wilhelm
Journal of Sleep Research September 9, 2026 DOI: 10.1111/jsr.70443 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Randomized, blinded, crossover laboratory study Peer reviewed |
|---|---|
| Sample size | 25 |
| Population | Adults experiencing at least one nightmare per week (21 female, mean age 23.12 years, SD 3.70) |
| Duration | 1-week deep breathing relaxation training; two laboratory nights |
| Measures | Beta (16.25-31 Hz) power, gamma (31.25-45 Hz) power, spindle count, spindle density, subjective wake-after-sleep-onset, nightmare symptoms |
| Topics | Dreaming |
| Key points | Odour-cued reactivation of a practiced deep-breathing relaxation exercise did not influence beta or gamma power or nightmare symptoms, but was unexpectedly associated with a significantly reduced spindle count and density. Spindle count during reactivation nights correlated positively with subjective wake-after-sleep-onset. The authors conclude that the intervention did not modulate cortical hyperarousal or nightmares, and that the functional significance of the spindle effect remains to be determined. |
Abstract
High-frequency EEG activity during sleep, reflecting cortical hyperarousal, has recently been incorporated into aetiology models of nightmares. However, its precise role remains insufficiently understood because experimental approaches to manipulate hyperarousal during sleep are limited. In the present study, we investigated whether re-presenting a relaxation-associated odour cue during sleep to reactivate a previously practiced deep-breathing relaxation exercise reduces cortical hyperarousal as indexed by beta (16.25-31 Hz) and gamma (31.25-45 Hz) activity and, consequently, nightmare symptoms. Twenty-five participants (21 female, mean age (SD) = 23.12 (3.70) years) experiencing at least one nightmare per week completed a 1-week deep breathing relaxation training paired with an odour cue. In a randomised, blinded, crossover design, either the relaxation-associated odour or a control odour was presented across two laboratory nights. Odour-based reactivation did not influence beta or gamma power but was unexpectedly associated with a significantly reduced spindle count and density. Spindle count during reactivation nights correlated positively with subjective wake-after-sleep-onset. Nightmare symptoms were unaffected. These findings suggest that odour-cued reactivation of a relaxation exercise did not modulate cortical hyperarousal or nightmare symptoms, despite altering spindle activity. The functional significance of this spindle effect remains to be determined. Future studies should employ longer reactivation periods and alternative interventions targeting cortical hyperarousal to further clarify its causal role in nightmare aetiology and the therapeutic potential of sleep-based interventions.