Music can improve sleep, even to the point that it is recommended to nurses for use in hospitals.
Another study analyzed a large body of existing research, and still found that music is effective at improving sleep, and can be clinically useful.
Even people with clinical insomnia can benefit from listening to music before sleep.
Pzizz also uses progressive relaxation, and in our latest survey of research we’ve found a controlled study that looks at progressive relaxation as a way of improving sleep.
In this study, sleep scientists found that progressive relaxation was effective at improving sleep, more effective than the control, mainly by decreasing the time it took to fall asleep.
They also tested and confirmed what we already knew, that hypnotic induction and suggestion can help improve sleep.
What was nice about this test of hypnosis, is the way it showed that hypnosis and progressive relaxation decreased the time it took to fall asleep, whereas the previously mentioned study mainly focused on time spent in deep sleep.
Finally, this study showed that hypnotic induction and hypnosis combined were more effective at improving the quality of sleep than self-relaxation and no-intervention combined. This means that you are much better off using hypnosis and progressive relaxation than you are trying to relax yourself, or doing nothing at all.
Sleep researchers decided to contrast the effectiveness of hypnotic and progressive relaxation on sleep onset. Using a validated self-report questionnaire, they examined the impact of these two psychotherapeutic techniques on sleep onset (latency). Validation, in this case, means that the self-report measure they used had been compared to other more direct measures of sleep onset and had been found to be accurate.
To avoid placebo effects (the effect of the trial itself) and demand characteristics (the tendency for participants to try to meet the expectations of the researchers) this study included a self-relaxation condition and a no-intervention condition. To further avoid any confounding of the effect of the individual conditions with the effect of the study as a whole, all participants were told that, while the study aimed at improving sleep, the real focus was to examine physiological response and the effect of arousal on outcome. This was both true and deceptive.
Based on the theory that insomnia is due, at least in part, to physiological arousal and tension, the researchers did measure arousal reduction (a.k.a. relaxation). However, informing the participants that the aim of the study was to examine physiology also served to distract them from the main focus, which was sleep onset. This kind of partial-deception is common in psychology experiments, where researchers are trying to avoid any interference from the participant’s understanding of the aim of study.
Psychology students at the University of Iowa were given a questionnaire that asked about sleep onset. Women who gave responses indicating trouble sleeping were then given a detailed follow up interview over the phone, which included questions about sleep related behavior and current life circumstances. Anyone using drugs, or receiving medical or psychological treatment was excluded from the study; this was to avoid confounding variables. The forty participants who met the selection criteria were then given a packet of forty sleep-questionnaires and told to immediately fill them out upon awakening each morning. Three participants were disqualified later on because they failed to show up to all the treatment sessions.
Female clinical psychology graduate students served as the therapists. They individually treated half of each condition. This was in an attempt to reduce the possibility of the therapist being a confounding variable. The therapist received extensive training from the lead researcher, and also followed detailed procedures.
Progressive relaxations involved the tensing and relaxing of different areas of the body, with subtle and indirect suggestions to relax. The hypnotic condition involved direct hypnotic suggestions to relax. Each participant underwent a session a week for three weeks. In the self-relaxation control condition, participants were instructed to focus on neutral imagery and the resulting bodily sensations. In the inactive control, participants believed they were on a wait list, and continued to fill out the questionnaire.
Measures were taken in an air-conditioned and sound-controlled room. Measures included heart rate, breathing rates (by measuring expansion and contraction of the abdomen), sweating (by measuring the degree to which the skin conducted electricity, with more conductance indicating more sweating), and muscle tension via electromyography (or the measuring of electrical activity in muscle tissue via the insertion of an electrode under the skin). Measures were scored for the last twenty seconds of the adaptation period, and the last twenty seconds of the intervention period. For the electromyography, the peak of each second was used as the data point for that second.
The sleep questionnaire contained questions about several aspect of sleep analysis, yet found the only significant effect was on sleep onset (latency). Statistical analysis examined the relationship between treatment condition and changes in sleep onset latency; both hypnotic and progressive relaxation conditions resulted in a significantly greater reduction in the time taken to fall asleep, as compared to the no-treatment control. All three active conditions were found to be significantly better than the waitlist condition. Post hoc analysis showed that hypnotic treatment combined with progressive relaxation was statistically significantly more effective than self-relaxation combined with no-treatment. The physiological measures showed a clear reduction of arousal during treatment. However, correlations between physiological scores and questionnaire scores were not strong. Progressive relaxation and hypnotic suggestion were equally effective at decreasing time before falling asleep, and were comparable when it came to the number of times participants woke up in the night, and their feelings of restfulness in the morning.
One limitation is that the sample size is too small to analyze the effect of the therapist, although the counterbalancing of therapists does help somewhat to alleviate this concern. Another limitation is the fact that the demand characteristics were not controlled for, although some amount of deception was employed to reduce the likelihood of this confounding the results. An EEG study, examining the actually brain-wave patterns would be more effective in avoiding this possibility of confound. This study does not elaborate on the differences between the active therapy conditions and the self-relaxation condition. The study does not include information about blinding. Finally, because this study is somewhat older, it uses methodology and sources that are outdated.
This study is published in a reputable journal with a high impact factor (a measure of the rate at which it is cited by other scientists). This study uses active controls, and deception to avoid the placebo effect and to reduce the likelihood of demand characteristics confounding the results. It also looks at changes over a period of three weeks, adding to the study’s validity.