Effects of preferred familiar music on falling asleep, The

effects of preferred familiar music on falling asleep, The

Iwaki, Tatsuya

The purpose of this study was to examine whether or not listening to music promotes falling asleep. Twenty university students, who usually listen to music at bedtime, were asked to take a nap in the laboratory while being monitored by a polysomnograph. Each participant selected preferred music to be played as they fell asleep. Stage 2 sleep latency was shorter for those sleeping with music playing compared to the no music control group. This tendency was reversed when participants tried to fall asleep quickly. Differences in sleep latency between the music and control conditions were due to the amount of episodic wakefulness. Results imply that music promotes or interferes with falling asleep by modulating the appearance of episodic wakefulness. Considering the mood, especially pleasantness while falling asleep and the ironic process theory of mental control (Wegner, 1994), the mechanisms of the effects of music on sleep are discussed. There is a possibility that listening to music promotes falling asleep though this may only be effective ofter balancing other factors.

No one knows how many people listen to music at bedtime. Many studies report that music is able to improve mood and help individuals cope with stress (Watkins, 1997). There is, therefore, a possibility that listening to music at bedtime can promote sleep.

Some studies have investigated the effects of music on sleep. Gitanjali (1998) confirmed that traditional Indian music, such as lullabies, had no effect on sleep. Rutter and Waring-Paynter (1992) reported that individuals that tend to sleep for a shorter amount of time (i.e., for fewer hours per night) were less satisfied with their sleep when music was playing. Using a polysomnograph, Sanchez and Bootzin (1985) compared the objective measures for sleep with music, white noise, and with no sound (control). Results of their study suggested that music increased the latency to sleep onset and wake time compared with white noise. Music did not promote sleep possibly because individuals attended to it. These studies did not support the positive effects of music on sleep. Moreover, Ansfield, Wegner, and Bowser (1996) explained insomnia using the theory based on the ironic process of mental control (see Wegner, 1994). This theory specifies conditions under which the desire to control a mental state (such as sleep, mood, attention, etc.) can yield the opposite of what is desired which is what seems to happen to chronic insomniacs. In terms of sleep onset, the urgent desire to fall asleep combined with a mental load should then, according to this theory, lead to wakefulness. The Sousa marches used in the Ansfield et al. study (1996) did not disturb those subjects who were not trying to sleep but were particularly bothersome to those who were trying to sleep. Results of the study imply that the relationship between music and sleep is not simple.

In a previous study (Iwaki, Tanaka, & Hori, 1999), approximately 15% of 304 Japanese university students indicated that they frequently listened to music at bedtime. It is unlikely that those people listened to music in order for it to interfere with sleep. In general, the effect of music is based on the interaction between an individual and a musical selection. Both these factors need to be considered when examining the effects of music on sleep. Studies where music tended to interfere with sleep (see above) did not take these factors into account. It seems likely that people who listen to music at bedtime on a regular basis would be the ones to ask about the features of musical selections that tend to promote sleep. Those who typically play music while they sleep might, however, be under the impression that music is effective in promoting sleep. There is a possibility that music actually delays the latency of sleep onset. It is therefore necessary to confirm the effects of music on sleep for those people using an objective index.

Moreover, if music is considered effective in promoting sleep, listening to music should be considered for the relief of insomnia. Most insomnia is psychophysiologic insomnia, a disorder of somatized tension causing physiologic arousal. When compared to normal sleepers, patients report a greater number of intrusive thought about sleep and report presleep cognitions as worried and negative (Borkovec, Lane, & VanOot, 1981). Considering that initial sleep onset difficulty seems to develop a self-perpetuating cycle of chronic insomnia, the hypnagogic period (transition period between sleep and awakening) should be noted. Traditional approaches have emphasized pharmacological treatment of insomnia. Holbrook, Crowther, Lotter, Cheng, and King (2000) reported that a meta-analysis of sleep records indicated that, when compared with placebo, benzodiazepines decreased sleep latency by 4.2 minutes. Recently, behavioral therapy has been considered the most appropriate treatment for patients with primary insomnia. Smith et al. (2002) compared the short-term efficacy of pharmacotherapy and behavioral therapy in primary insomnia using a meta-analysis. Results indicated that behavioral therapy resulted in a greater reduction in sleep latency than did pharmacotherapy. The therapies are based on the fact that primary insomnia is associated with physiologic, emotional, and cognitive arousal, and conditioning arousal before sleep onset. Music might be able to be used as a tool in helping behavior therapy.

The purpose of this study was to examine whether or not music promoted falling asleep. In order to help confirm a possibility that music promotes falling asleep, subject selection was limited to those who typically had music playing during sleep as well as their preferred musical selections. This was considered to be the optimum condition to analyze the effects of music on falling asleep. Considering its application to insomnia (see Ansfield et al., 1996) an attempted sleep condition was also set up in order to focus on the hypnagogic period.


Participants and Music

Twenty healthy university students (male = 10, female = 10, 20-28 years in age) participated in this experiment. They reported that they usually listened to music at bedtime. Participants were divided into two groups (natural sleep group, n = 10; attempted sleep group, n = 10). All participants were informed about the study protocol and agreed with the procedures. The music used for the study was a series of pieces chosen from each subject’s preferred music album.


Participants were studied in a 3 m x 3 m electrically-shielded, sound-attenuated, and air conditioned bedroom. Temperature readings taken for the duration of the study rendered a mean of 22 +/-1.0 deg C.

The apparatus used to play the music were a DENON DCD-1600 CD player, a YAMAHA AX-10 amplifier, and a DIATONE DS-60025 speaker system. The distance between the bed and the two speakers was 1.5 m. The sound pressure level was adjusted at a comfortable listening volume for each participant (about 55-65 dB(A)).

All electrophysiological parameters were recorded simultaneously both on a conventional polysomnographic hardcopy (NEC San-Ei 1A57) and a digital data recorder (TEAC DR-M3). Surface electrodes were placed on two central scalp areas (C3 & C4) of the international 10/20 system referenced to ipsilateral ear lobes (EEG), two horizontal electro-oculograms (EOG), each referenced to ipsilateral ear lobes for slow eye movement, on the chin (mentalis electromyogram, EMG), referenced to each other, and on the forehead for the body ground. All electrophysiological parameters were recorded using silver-silver chloride disk electrodes. Interelectrode impedance was below 5 K(Omega) The EEGs were amplified using a high-cut filter setting of 60 Hz and time constant of 0.3 seconds.


Participants performed a logic task of 5 min. after changing into their nightclothes and attaching electrodes. This period was designed to lead the participants to full awake. After task performance, participants lay down on the bed and filled out a questionnaire about sleepiness. Participants were instructed to close their eyes after the light was turned off. In the attempted sleep group, participants were asked to fall asleep “as fast as you can” while participants in the natural sleep group were instructed to fall asleep “whenever you want.” After the lights were turned off, the music each participant had recommended was played (music condition). An experimenter woke the participant 5 minutes after the appearance of the first sleep spindle by calling his/her name through an intercom system before entering the room. The experiment was started between 14:00 and 15:00 when the participant was likely to become sleepy.

After a 1-week interval, participants fell asleep under the same orders (as fast as you can or whenever you want) but no music was played (control condition). All participants did fall asleep during both the music and control conditions. The order of experimental conditions was alternated from participant to participant.


Subjective variables were sleepiness, pleasantness, and subjective estimate of sleep latency. Sleepiness before falling asleep was evaluated using the Kwanseigakuin Sleepiness Scale (KSS), a 22-item check list. This scale was based on the Stanford Sleepiness Scale using Thurstone’s method of equal-appearing intervals (Ishihara, Saito, & Miyata, 1981). Pleasantness was evaluated using a 7-point scale of a semantic differential type instrument by which participants rated the mood they had experienced during the previous bedtime. Subjective sleep latency was also estimated a number of minutes after awaking.

Sleep variables were sleep latency and the amount of time of each sleep stage. Sleep records were scored visually in 10 s epochs according to standardized criteria (Reftschaffen & Kales, 1968) for Awake, Stage 1 sleep, and Stage 2 sleep (see Figure 1). Stage 1 sleep was also referred to as drowsiness or presleep and is the first or earliest stage of sleep. Stage 1 sleep was scored when the alpha activity in the EEG dropped to less than 50%. The EOG showed slow, rolling eye movements, especially early in the stage. Stage 2 sleep was the predominant sleep stage during a normal night’s sleep. The distinct and principal EEG criterion to establish Stage 2 sleep was the appearance of sleep spindles. These are paroxysms of 12-14 Hz activity persisting for at least 0.5 s. Sleep latency measurements were determined by the first epoch in Stage I and in Stage 2. Appearance time of Awake and Stage 1 sleep were also measured by the amount of time for each stage until it reached stage 2 sleep. Appearance time was not necessarily in accordance with sleep latency because it is common for a hypnagogic period to return to Awake even once in Stage 1 sleep. Each record was scored entirely by one scorer and was re-scored by an independent scorer. A comparison was then made between the two scoring on an epoch-by– epoch basis to determine between-scorer reliability which exceeded 90% for each participant’s record. Subjective and sleep variables were analyzed using two-way ANOVAs (group: natural vs. attempted sleep x condition: control vs. music).


Sleepiness, Pleasantness, and Self-estimation of Sleep Latency

Table 1 shows the mean ratings and standard deviations for each group and each condition in sleepiness, pleasantness, and subjective estimate of sleep latency. For sleepiness, there were no significant effects for group, F(1, 18) = .03, ns, condition, F(1, 18) = .06, ns, or group by condition interactions, F(1, 18) = 1.27, ns. It was confirmed that sleepiness was not different between groups and/or conditions.

For pleasantness, the effect for group by condition interaction was significant, F(1, 18) = 5.56, p

For self-estimation of minutes of sleep latency, there were no significant effects for group, F(1, 18) = .00, ns, condition F(1, 18) = .47, ns, or for group by condition interaction, F(1, 18) = .99, ns. It was confirmed that estimated sleep latency was not different between groups and/or conditions.

Latency to Stage 1 Sleep and Stage 2 Sleep

Figure 2 shows the mean latencies and standard deviations for each group and condition in Stage 1 sleep and Stage 2 sleep. The mean number of minutes to Stage 1 sleep was slightly less in the music condition compared with the no music control condition for the natural sleep group while it increased in the attempted sleep group. This tendency was emphasized in the latency of Stage 2 sleep. For Stage 1 latency, no significant effects were found for group, F(l, 18) = .00, ns, condition, F(1, 18) = .44, ns, or for group by condition interaction, F(1, 18) = 2.21, ns. For Stage 2 latency, the effect for group by condition interaction was significant, F(1, 18) = 4.93, p

Appearance Time o fAwake and Stage 1 Sleep

Figure 3 shows the mean number of minutes and standard deviations for each group and condition in Awake and Stage I sleep. Although there was less awake time during the music condition than in the no-music control condition for the natural sleep group, there was more awake time in the music condition than in the no– music attempted sleep group. These tendencies were not found in the appearance time of Stage 1 sleep. For the amount of awake, the effect for group by condition interaction was significant, F(1, 18) = 5.62, p


The purpose of this study was to examine whether or not music promoted falling asleep. Participants chosen for the study reported that they usually listened to music at bedtime.

Results of the latency to Stage 2 sleep suggested that music did promote falling asleep in the natural sleep group (fall asleep whenever you want) whereas music interfered with falling asleep in the attempted sleep group (fall asleep as fast as you can). Moreover, it was found that the effects of music on Stage 2 sleep latency depended on the appearance time of “awake.” It is suggested that this “awake” is episodic wakefulness, which was generated once it reached Stage 1. In the hypnagogic period, sleep deepens gradually with repeated rising and falling wakefulness (Tanaka, Hayashi, & Hori, 1996). Findings in the present study suggest that the music controls and promotes episodic wakefulness.

In the natural sleep group, it was found that music both controlled the episodic wakefulness and helped progress falling asleep. For example, deep sleep occurs quickly after sleep deprivation and there is poor episodic wakefulness in the hypnagogic period (Ogilvie & Wilkinson, 1984). It is, however, difficult to understand that the decrease in the episodic wakefulness as a result of listening to music functioned directly on the wake/sleep mechanism similar to sleep deprivation. On the other hand, since the feature of the hypnagogic period is an increment of the sensory stimulus threshold (see Bonnet, 1982), the psychological influence accompanying listening to music (e.g., imagery experience and change in mood) might control the attention to the outside environment and set better conditions to help progress to falling asleep more promptly. Browman and Tepas (1976) reported that the latency of sleep onset was shorter after a brief relaxation period before bedtime than after light dynamic exercise and a boring/monotonous vigilance task. The difference in pleasantness between the control and music conditions in the natural sleep group might suggest that music plays a role in adjusting mood prior to falling asleep. Considering the positive correlation between anxiety ratings and the number of wakings (Rosa, Bonnet, & Kramer, 1983), if music reduced anxiety levels even temporarily, there is a possibility that music would promote falling asleep. In the present study, each participant selected preferred music to be played as they fell asleep. Music selected by subjects/patients has reduced anxiety and tension before surgery and during task performance (see Allen & Blascovich, 1994; Davis & Thaut, 1989; Thaut & Davis, 1993).

Although participants and selected music were selected as the same criterion in the natural sleep group, listening to music in the attempted sleep group increased the episodic wakefulness and delayed sleep onset when compared to the no-music presentation. These results support the findings that most subjects chose not to listen to music when they failed to fall asleep (Iwaki et al., 1999). Moreover, these observations confirm the results of a previous study (Ansfield et al., 1996) when considering the physiological aspect. In the Ansfield et al. study, subjects listened to high cognitive load music while trying to sleep resulting in a longer sleep latency than those who were not trying to sleep. According to the theory of the ironic process of mental control (Wegner, 1994), the operation process that requires greater cognitive capacity promotes falling asleep by searching for mental content consistent with the intended state. When music listening consumed the cognitive capacity, the operating process might be disturbed and thus enhance the person’s sensitivity to the mental content when they are not able to fall asleep.

As far as the attempted sleep group is concerned, the latency to Stage 2 sleep in the control condition was similar to that of the music condition in the natural sleep group. This reflects the success of the intention to fall asleep. Furthermore, the latency to Stage 2 sleep in the music condition was similar to that of the control condition in the natural sleep group. Music does not necessarily interfere with falling asleep. Based on the theory of the ironic process (Wegner, 1994), these results may be due to the reduction of the cognitive load resulting from using music which was familiar to the participants. If the stimuli are novel, a strong cognitive effort would likely be needed (see Ohman, 1979).

Results of the present study suggest that music does promote falling asleep for those who are used to listening to music during bedtime. There is a possibility that music could be useful for sleep problems as a nonpharmacological approach (Hanser, 1990). It is suggested, however, that this effect is realized after a delicate balance between specific factors is achieved. These factors include the adjustment of one’s mood as it relates to the preferred music, the information load resulting from the interaction between the music and intention to sleep, and the skill of listening to music for falling asleep. Within the framework of the effect process of music on falling asleep, it is not clear if music promotes falling asleep or interferes with it. Further study regarding the mechanisms which music listening affects while falling asleep is required.


Allen, K., & Blascovich, J. (1994). Effects of music on cardiovascular reactivity among surgeons. Journal of the American Medical Association, 272, 882-884.

Ansfield, M. E., Wegner, D. M., & Bowser, R. (1996). Ironic effects of sleep urgency. Behavioral Research and Therapy, 34, 523-531.

Bonnet, M. (1982). Performance during sleep. In W. B. Webb (Ed.), Biological rhythms, sleep, and performance (pp. 205-237). New York: John Wiley & Sons, Ltd. Borkovec, T. D., Lane, T. W., & VanOot, P. H. (1981). Phenomenology of sleep

among insomniacs and good sleepers: Wakefulness experience when cortically asleep. Journal of Abnormal Psychology, 90, 607-609.

Browman, C. P., & Tepas, D. 1. (1976). The effects of presleep activity on all-night sleep. Psychophysiology, 13, 536-540.

Davis, W. D., & Thaut, M. H. (1989). The influence of preferred relaxing music on measures of state anxiety, relaxation, and physiological responses. Journal of Music Therapy, 26, 104-114.

Gitanjali, B. (1998). Effect of the Karnatic music raga “Neelambari” on sleep architecture. Indian Journal of Physiology and Pharmacology, 42, 119-122.

Hanser, S. B. (1990). A music therapy strategy for depressed older adults in the community. Journal of Applied Gerontology, 9, 283-298.

Holbrook, A. M., Crowther, R., Lotter, A., Cheng, C., & King, D. (2000). Meta-analysis of benzodiazepine use in the treatment of insomnia. Canadian Medical Association Journal, 162, 225-233.

Ishihara, K., Saito, T., & Miyata, Y (1981). Sleepiness scale and an experimental approach. The Japanese Journal of Psychology, 52, 362-365.

Iwaki, T., Tanaka, H., & Hori, T. (1999). An exploratory survey of using music for going to sleep. TheJournal of Japan Biomusic Association, 17, 200-209.

Ogilvie, R. D., & Wilkinson, R. T. (1984). The detection of sleep onset: Behavioral and physiological convergence. Psychophysiology, 21, 510-520.

Ohman, A. (1979). The orienting response, attention, and learning: An information processing perspective. In H. D. Kimmel, E. H. van Olst, &J. F. Orlebeke (Eds.), The orienting reflex in humans (pp. 443-471). Hillsdale, NJ: Lawrence Erlbaum.

Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology, techniques, and scoring systems for sleep stages of human subjects. Washington, DC: Public Health, US Government Publishing Office.

Rosa, R. R., Bonnet, M. H., & Kramer, M. (1983). The relationship of sleep and anxiety in anxious subjects. Biological Psychology, 16, 119-126.

Rutter, S., & Waring-Paynter, K. (1992). Prebedtime activity and sleep satisfaction of short and long sleepers. Perceptual and Motor Skills, 75, 122.

Sanchez, R., & Bootzin, R. R. (1985). A comparison of white noise and music: Effects of predictable and unpredictable sounds on sleep. Sleep Research, 14, 121. Smith, M. T., Perlis, M. L., Park, A., Smith, M. S., Pennington, J., Giles, D. E., &

Buysse, D. J. (2002). Comparative meta-analysis of pharinacotherapy and behavior therapy for persistent insomnia. American fournal ofPsychiatry, 159, 5-11. Tanaka, H., Hayashi, M., & Hori, T. (1996). Statistical features of hypnagogic EEG measured by a new scored system. Sleep, 19, 731-738.

Thaut, M. H., & Davis, W. D. (1993). The influence of subject-selected versus experimenter-chosen music on affect, anxiety, and relaxation. Journal of Music Therapy, 30, 210-223.

Watkins, G. R. (1997). Music therapy: Proposed physiological mechanisms and clinical implications. Clinical Nurse Specialist, 11, 43-50.

Wegner, D. M. (1994). Ironic process of mental control. Psychological Review, 101, 34-52.

Tatsuya Iwaki, PhD

Hideki Tanaka, PhD

Hiroshima International University

Tadao Hod, PhD

Hiroshima University

Copyright American Music Therapy Association Spring 2003

Provided by ProQuest Information and Learning Company. All rights Reserved