A Thermal Environment that Promotes Efficient Napping Miki Nakai,1 Tomoyoshi Ashikaga,2 Takahiro Ohga,3 Keiki Takadama4 DAIKIN INDUSTRES LTD., 1,2,3 The University of Electro-Communications,4 miki2.nakai@daikin.co.jp,1 tomoyoshi.ashikaga@daikin.co.jp,2 takahiro.Ohga@daikin.co.jp,3 keiki@inf.uec.ac.jp4 Abstract the nap (eliminating drowsiness and improving perfor- People need to maintain good health to live productive and mance) is not clear. In previous studies about nap, it has creative lives, and sleep greatly contributes to maintaining been described about optical nap length (Hayashi et al., good health. In recent years, some companies have begun al- 1999, 2003, 2004, 2005), light effect (Hayashi et al., 2007), lowing time for sleep during work hours as “power naps” to sound effect (Toma et al., 2004). However, it is still un- eliminate drowsiness and improve efficiency. In this article, known about effect of a thermal environment on nap qual- we report on the optimal thermal environment for efficient ity. naps during the daytime when the body is active and the ef- Since the environmental temperature controlled by the air fect that naps have on improving productivity. This study uses questionnaires on drowsiness before and after the test conditioner directly affects the body temperature regula- along with the Psychomotor Vigilance Task (PVT) and N- tion, it is considered to have a great influence on the "sleep back task to evaluate productivity improvements in test sub- quality" that rests the brain. In addition, air conditioners are jects when room temperature is changed for each scene of becoming widespread all over the world, and more and “reclining for nap,” “sleeping,” and “waking from nap.” Test more homes are installing "one air conditioner in one results show that differences in room temperature can pro- room". Therefore, in this study, we focus on thermal con- mote the onset of sleep, maintain sleep, and stimulate wake- trol during nap using an air conditioner that can be widely fulness. By optimally controlling these factors, it is possible applied, and report the results of evaluating the effect of to take a short high-quality nap to improve productivity. As environmental temperature control on the quality of nap sleep is said to be related to the function of the autonomic (eliminating drowsiness and improving performance). nervous system, it is believed that the thermal environment exerts influence on the autonomic nervous system and affects the quality of sleep, even when sleep duration is short. 2. Optimal thermal control for daytime naps 1. Introduction Room temperature inside the nap room is adjusted during the 30-minute nap according to the three stages of “Reclin- Recently, the amount of sleep that many people in Japan ing for nap,” “Sleeping,” and “Waking from nap.” And the obtain each night has declined (NHK Broadcasting Culture environmental temperature at each stage was controlled to Research Institute, 2020), and more than 70% of Japanese ensure a comfortable and moderate sleep depth. Fig 1 show men and women 20 years of age and older sleep an average the example of the thermal control for nap. of less than 7 hours. Consequently, more than 30% of Jap- anese people between the ages of 20 and 59 years old suffer (1) Reclining for nap... from “daytime drowsiness” at least three times a week The warm environment is thought to enhance relaxation (Ministry of Health, Japan, 2019), leading to a decline in before the nap, which accelerate the onset of sleep. productivity. So, there is increasing interest in good sleep. (2) Sleeping... There are the following previous studies on nighttime sleep Assist restful sleep with a Slightly cooler room temperature and athletic performance (Cheri D Mah et al., 2011). We (3) Waking from nap... measured the shooting accuracy, evels of daytime sleepi- At higher temperatures, it encourages the body to awaken ness, and Psychomotor Vigilance Task (PVT) when 10 basketball players slept for 10 hours or more and did not. And Fig.2 show the typical example of sleep depth by this As a result, when sleeping for 10 hours or more, the shoot- thermal control (W:wake, R:REM sleep, N1: NREM Sleep ing accuracy was significantly improved and the drowsi- Stage 1, N2: NREM Sleep Stage 2, N3: NREM Sleep Stage ness and PVT reaction time were decreased as compared with the short sleep. 3). It is said that naps are also effective in relieving drowsi- ness and improving performance. But the effect of the sur- rounding environment when taking a nap on the quality of ___________________________________ In T. Kido, K. Takadama (Eds.), Proceedings of the AAAI 2022 Spring Symposium “How Fair is Fair? Achieving Wellbeing AI”, Stanford University, Palo Alto, California, USA, March 21–23, 2022. Copyright © 2022 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). 88 3.2 Experimental equipment Test subjects wear an electroencephalograph G (Sleep Scope, SleepWell Co., Osaka, Japan) (Yoshida et al., 2015) and EEG during sleep of the subject was measured.And The room temperature of the nap room was measured using en- vironmental sensor (2JCIE-BU01, OMRON Co., Kyoto, Ja- pan). Fig. 1 the thermal control example for nap 4. Method 4.1 Participants and Protocol The test subjects are 6 males and 5 females in their twenties, and the amount of their clothes was kept constant. From Jan- uary 2020 to February 2022, 143 tests were conducted. The test subject is given a questionnaire on drowsiness (The Stanford Sleepiness Scale, Table1), and 2 types of produc- Fig. 2 the typical example of sleep depth tivity measurement tests (PVT-B; Psychomoter Vigilance Task, Fig. 4 and N-Back task, Fig. 5) are conducted. PVT- B objectively assesses fatigue-related changes in alertness nap associated with sleep loss, extended wakefulness, circa- dian misalignment, and time on task. In this nap test, the re- 3. Experimental Environment action speed was measured by clicking a PC mouse in re- 3.1 nap room sponse to a lit signal on the PC monitor in order to evaluate A nap room of H2315 x W (2400 + 385) x D1200 mm (Fig. awakening degree. N-Back task is used extensively in liter- 3) was created, and a bed was installed in the nap room. Dur- ature as a working memory (WM) paradigm. In this nap test, ing the test, the entrance sliding door was kept closed, and the test subjects have to memorize details to several ques- the test was conducted while ensuring sound insulation and tions while answering a simple calculation formula to eval- light shielding. The humidity inside the experimental nap uate it. room remains within a range of 40-60%, and a small space After that, a 10-minute rest period is provided before the multi-cassette air conditioner "cocotas" manufactured by test subject takes a 30-minute nap. After waking, the test Daikin Industries was installed to control the temperature of subject receives another rest period of 5 minutes and is again the nap room. given a questionnaire on drowsiness. These productivity measurement tests are administered twice to the test subject after waking at 1-hour and 3-hour intervals. Table 1 The Stanford Sleepiness Scale Fig. 3 nap room 89 Fig. 4 PVT-B screen Fig. 6 the thermal control example for nap (A: the temperature control in a V-shaped pattern, B:no thermal control) Table 2 Evaluation items and contents Fig. 5 N-Back Task screen 4.2 Room temperature conditions Room temperature inside the nap room is adjusted during the 30-minute nap according to the three stages of “Reclin- 5. Results ing for nap,” “Sleeping,” and “Waking from nap” as shown 5.1 (1) Reclining for nap in Fig.6(A,B). By room temperature, it is measured about Fig. 7 shows the results of sleep onset latency at room "changes in sleep depth" and evaluated about "productivity temperature before falling asleep. In order to realize a ther- before and after the nap" as shown in Table2. At the time of mal environment that encourages rapid sleep onset, we hy- nap, the following two room temperature environments pothesized that by setting the room temperature before fall- ing asleep to a warmer temperature, the relaxing effect were applied. would be enhanced and it would be easier to fall asleep. A: the temperature control in a V-shaped pattern Room temperature before the onset of sleep is set at one of -It is kept slightly warm room temperatures before sleep three settings: low (25 ℃ or less), neutral (26 ℃), or high onset, kept lower after sleep onset, and raised before waking. (27℃). The time required before the onset of sleep at each B: no thermal control temperature environment is evaluated. As expected, A ten- dency was seen in which sleep onset latency shortened -It is maintained a neutral temperature from before sleep when the room temperature was kept slightly warm (27℃). onset to waking up. 90 5.1 (3) Waking from nap Contrary to the sleeping con- ditions during sleep, we hy- pothesized that "a high room temperature has the effect of promoting awakening" and verified that raising the room temperature immediately be- fore waking up promotes awakening with less drowsi- ness. We compared the differ- ence in sleep depth between the case where the room tem- Fig. 9 Sleep depth percent- perature was not changed dur- age (%) and room tempera- Fig. 7 Sleep onset latency (minutes) and room temper- ing sleep and the case where ture (° C) before waking ature (℃) before sleep onset the room temperature was raised to 27 ° C or higher 3 minutes before the wake-up time. 5.1 (2) Sleeping As a result, as shown in Fig. 9, when the room temperature Daytime naps are said to be "optimal for NREM Sleep 3 minutes before waking up was 27 ° C or higher, the sleep Stage 2 (N2)", which does not increase sleep inertia after depth tended to be shallower. waking up (Stampi C. et al., 1990). Therefore, with the aim of realizing a "thermal environment that can maintain N2" during sleep, the hypothesis that "a slightly cooler tempera- 5.1 Result: Thermal control method obtained from the ture is comfortable and does not interfere with sleep", that demonstration result is, "a thermal environment that can maintain sleep". The ex- From the results of 5.1(1), (2) and (3), the thermal control periment was conducted. It was confirmed that test subjects method shown in Fig. 10 is considered to be optimal for ef- reached and maintained N2 10 minutes after the onset of ficient short-time sleep during the day. sleep at a room temperature of 26℃, compared to no change in room temperature (Fig.8 (A)-1, (B)-1). Fig. 8 (A) -2 and (B) -2 show examples of sleep depth when the test was per- formed with control A or B. Fig. 10 optimal thermal control 5.2 Performance improvement effect with or without thermal control This time, we also verified the effect of the application of thermal control on the performance after waking up. As for the verification method, three items of "reaction speed, working memory, and drowsiness" were compared 1 hour after waking up and 3 hours after waking up. As a result, in Fig. 8 Sleep depth percentage (%) and elapsed time from about 50 cases tested with thermal control, which provides sleep (in minutes) (A-1: control A, B-1: control B) the best sleep condition, performance was improved com- Sleep depth and elapsed time from sleep (in minutes) pared to the case without thermal control. The results are (A-2: control A, B-2: Control B) shown in the Table 3. 91 Table 3 Improvement in reaction speed, working 6.2 Circadian rhythm memory, and drowsiness by thermal control compared Under the influence of circadian rhythm, it is generally said to cases without thermal control that it is hard to feel sleepy in the morning and it feels sleepy in the afternoon (Lavie, P et al., 1985). 6. Discussions We introduced a thermal control that allows you to take a nap comfortably and effectively even during the daytime when your arousal level is high. In addition, taking a 30- Fig. 11 Sleep onset la- Fig. 12 Sleep onset la- minute nap during the day using this technology has the ef- tency (minutes) for men tency (minutes) for men fect of reducing drowsiness after waking up and improving and women at room tem- and women at room reaction speed and working memory compared to taking a perature of 25° C temperature of 27 ° C nap without thermal control. In future efforts, we would like to consider customizing the thermal control in consideration of the following two points. Fig.13 shows an example of general sleep depth when tak- 1). Gender difference ing a nap in the morning and taking a nap in the afternoon. 2). Circadian rhythm As these figures show, drowsiness is less likely to be felt in In fact, the results of this experiment also show the effects the morning, resulting in lighter sleep and increased awak- of 1). gender differences and 2). circadian rhythms, and we ening. On the other hand, since it is easy to feel drowsiness believe that more precise evaluation of these results will en- in the afternoon, the depth of sleep tends to become deeper able the realization of individual thermal control methods. as time passes after sleep onset. Therefore, by applying the result of 5.1(3), in the case of a nap in the morning, we will consider maintaining the room 6.1 Gender difference temperature while sleeping without raising it before waking Regarding 5.1 (1), we received multiple answers such as up in order to obtain sufficient N2 time. In the case of a nap "it was cold" and "it was hot" in the questionnaire after in the afternoon, we will consider raising the room temper- taking a nap, so we examined how to feel the temperature ature before waking up in order to avoid reaching N3. between men and women. We compared sleep onset latency by gender for the lower room temperature of 25°C. As a re- sult, as shown in Fig.11, no significant difference was ob- (a) (b) served between men and women. However, 43% of women Sleep depth (3 out of 7 cases) and 23% of men (3 out of 13 cases) an- swered that they were "cold" by the time they fell asleep. Therefore, it was considered that women tended to feel colder. elapsed time from sleep (in minutes) On the other hand, when the room temperature was slightly warmer at 27 ° C, a significant difference was found when Fig. 13 Sleep depth and elapsed time from sleep (in comparing sleep onset latency by gender (Fig. 12, p <0.05). minutes) (a: in the morning, b: in the afternoon) Furthermore, 2.3% of women (1 out of 43 cases) and 5.1% of men (4 out of 78 cases) answered that they were "hot" by the time they fell asleep. Therefore, it was suggested that men may tend to feel hotter. 7. Conclusion Thermal control was introduced that enables comfortable and efficient naps even during the day when the degree of alertness is elevated. Additionally, when a 30-minute nap is taken during the day with heat control using this technology, 92 compared to no use of heat control, drowsiness decreases Yoshida, M.; Kashiwagi, K.; Kadotani, H.; Yamamoto, K.; while reaction speed and working memory improve. Koike, S.; Matsuo, M.; Yamada, N.; Okawa, M.; and Urade, Y. Validation of a portable single-channel EEG monitoring system. 2015. Journal of Oral Sleep Medicine, 1, 140–147. References Cheri, D.; Kenneth, E.; Eric, J.; and William, C.; 2011. The effects of sleep extension on the athletic performance of col- legiate basketball players. Sleep, 34(7):943-50 Hayashi, M.; Watanabe, M.; and Hori, T.; 1999. The effects of a 20-min nap in the mid-afternoon on mood, performance and EEG activity. Clinical Neurophysiology, 110, 272-279. Hayashi, M.; Fukushima, H.; and Hori, T.; 2003. The effects of short daytime naps for five consecutive days. Sleep Re- search Online, 5, 13-17. Hayashi, M.; Chikazawa, Y.; and Hori, T.; 2004. Short nap versus short rest: recuperative effects during VDT work. Er- gonomics, 47, 1549-1560. Hayashi, M.; Motoyoshi, N.; and Hori, T.; 2005. Recupera- tive power of a short daytime nap with or without stage 2 sleep. Sleep, 28, 829-836. Hayashi, M.; and Hori, T.; 2007. A short nap as a counter- measure against afternoon sleepiness, Japanese Journal of Physiological Psychology and Psychophysiology, 25(1): 45- 59. Lavie, P.; and Schulz, H.; 1985. Ultradian rhythms: gates of sleep and wakefulness., Ultradian rhythms in physiology and behavior. In Schulz, H; and Lavie, P.; (Eds.), Berlin: Springer-Verlag, 148–164. NHK Broadcasting Culture Research Institute, 2020. Na- tional Daily Life Survey Ministry of Health, Labour and Welfare, 2019. National Health and Nutrition Survey Stampi, C., Mullington, J.; Rivers, M.; Campos, J. P.; and Broughton, R.; 1990. Ultrashort sleep schedules: Sleep ar- chitecture and recuperative value of 80-, 50- and 20-min naps. In J. Horne (Ed.), Sleep'90. Bochum: Pontenagel Press, 71-74. Toma, A.; and Ogata, S.; 2004. Fundamental research to- ward the education practice which applied music: con- sciousness change on EEG under the music appreciation and mental set. The bulletin of the Research and Clinical Center for Handicapped Children, 6, 41-54 93