Night watch during REM sleep for the first-night effect

We experience disturbed sleep in a new place, and this effect is known as the first-night effect (FNE) in sleep research. We previously demonstrated that the FNE was associated with a protective night-watch system during NREM sleep in one hemisphere, which is shown as interhemispheric asymmetry in sleep depth in the default-mode network (DMN), and interhemispheric asymmetry in increased vigilance to monitor external stimuli. The present study investigated whether rapid eye movement (REM) sleep exhibited a form similar to a night-watch system during NREM sleep. First, we tested whether theta activity, which is an index of the depth of REM sleep, showed interhemispheric asymmetry in association with the FNE, by source-localizing to the DMN. However, interhemispheric asymmetry in theta activity during REM sleep was not found in association with the FNE. Next, we tested whether vigilance, as measured by evoked brain responses to deviant sounds, was increased in one hemisphere and showed interhemispheric asymmetry in association with the FNE during REM sleep. Because vigilance is different between the phasic period where rapid eye movements occur and the tonic period where rapid eye movements do not occur during REM sleep, REM sleep was split into phasic and tonic periods for measurements of evoked brain responses. While the evoked brain responses are generally small during the phasic period without the FNE, we found that the evoked brain response was significantly augmented by the FNE during the phasic period. In contrast, the evoked brain response during the tonic period did not differ by the presence of the FNE. Interhemispheric asymmetry in brain responses was not found during the phasic or tonic periods. These results suggest that a night-watch system for the FNE appears as interhemispheric asymmetry in sleep depth and vigilance during NREM sleep, but it appears as increased vigilance in both hemispheres during the phasic period, when vigilance to external stimuli is generally reduced without the FNE, during REM sleep. Therefore, a night-watch system associated with the FNE may be subserved by different neural mechanisms during NREM and REM sleep.

tested whether theta activity, which is an index of the depth of REM sleep, showed 23 interhemispheric asymmetry in association with the FNE, by source-localizing to the 24 DMN. However, interhemispheric asymmetry in theta activity during REM sleep was not 25 found in association with the FNE. Next, we tested whether vigilance, as measured by 26 evoked brain responses to deviant sounds, was increased in one hemisphere and showed 27 interhemispheric asymmetry in association with the FNE during REM sleep. Because 28 vigilance is different between the phasic period where rapid eye movements occur and 29 the tonic period where rapid eye movements do not occur during REM sleep, REM sleep 30 was split into phasic and tonic periods for measurements of evoked brain responses. 31 While the evoked brain responses are generally small during the phasic period without 32 the FNE, we found that the evoked brain response was significantly augmented by the 33 FNE during the phasic period. In contrast, the evoked brain response during the tonic 34 period did not differ by the presence of the FNE. Interhemispheric asymmetry in brain 35 responses was not found during the phasic or tonic periods. These results suggest that a 36 night-watch system for the FNE appears as interhemispheric asymmetry in sleep depth 37 and vigilance during NREM sleep, but it appears as increased vigilance in both 38 hemispheres during the phasic period, when vigilance to external stimuli is generally 39 reduced without the FNE, during REM sleep. Therefore, a night-watch system associated 40

Introduction 50
Sleep is crucial for the maintenance of daily life (Stickgold, 2005; Imeri and Opp, 51 2009). The psychological and behavioral consequences of declines in sleep quality may 52 be severe (Carskadon and Dement, 1981;Drummond et al., 2000;Buysse et al., 2011;53 Okawa, 2011;Czeisler, 2013). However, sleep may be decreased as a protective 54 mechanism under specific circumstances (Rattenborg et al., 1999;Peever and Fuller, 55 2017). One form of this mechanism is the first-night effect (FNE), which is widely 56 known in human sleep research (Agnew et al., 1966;Carskadon and Dement, 1979; 57 Tamaki et al., 2005a;Tamaki et al., 2005b;Tamaki et al., 2014;2016). The FNE is a 58 temporary sleep disturbance that occurs specifically in the first session of sleep 59 experiments in young healthy adults, and it manifests as prolonged sleep-onset latency, 60 frequent arousals, and decreased deep non-rapid eye movement (NREM) sleep (Roth et 61 al., 2005). Our previous study demonstrated that the FNE was not merely a sleep 62 disturbance but a manifestation of a protective night-watch system in one hemisphere, 63 which is less asleep and more vigilant than the other hemisphere to monitor unfamiliar 64 surroundings during deep NREM sleep (Tamaki et al., 2016). More specifically, slow-65 wave activity, which is an index of sleep depth during NREM sleep, decreases in the left 66 hemisphere compared to the right hemisphere regionally in the default-mode network 67 (DMN) on day 1 with the FNE, which causes interhemispheric asymmetry in slow-wave 68 activity. The amplitude of an evoked brain potential during deep NREM sleep correlates 69 with vigilance, and it was increased in one hemisphere on day 1, which caused 70 interhemispheric asymmetry in vigilance. These interhemispheric asymmetries in local 71 sleep depth and vigilance to monitor the external world function as a night-watch system, 72 which may counteract vulnerability during deep NREM sleep (Tamaki et al., 2016). 73 In contrast to NREM sleep, whether the FNE influences brain activities during 74  (Toussaint et al., 1997;Curcio et al., 2004). One study showed that the 83 FNE decreased electroencephalography (EEG) power in wide ranges of frequency bands, 84 including delta (0.5-3.5 Hz), theta (4-7.5 Hz), and beta (13-21.5 Hz) bands (Toussaint et 85 al., 1997). The other study reported that the FNE increased theta power (5 Hz) (Curcio et 86 al., 2004). No study investigated brain activities separately for each hemisphere or 87 whether the FNE altered vigilance during REM sleep. Therefore, whether the FNE alters 88 sleep depth or vigilance level during REM sleep to act as a night-watch system in an 89 analogous manner to NREM sleep is not known. 90 Notably, it is suggested that the ability to process external stimuli differs 91 depending on whether rapid eye movements appear (phasic period) or not (tonic period) 92 during REM sleep (Price and Kremen, 1980;Sallinen et al., 1996;Takahara et al., 2002;93 2006b; Wehrle et al., 2007). First, the amplitudes of evoked brain responses differ 94 significantly between the phasic and tonic periods. That is, much smaller brain responses 95 are observed during the phasic than the tonic period (Sallinen et al., 1996;Takahara et al., 96 2002;2006b). The amplitudes of evoked potentials correlate with the degree of vigilance 97 (Nielsen-Bohlman et al., 1991; Michida et al., 2005). Therefore, the finding of smaller 98 potentials during the phasic period suggests that the degree of vigilance to external 99 stimuli is significantly decreased when the eyes are moving during the phasic compared 100 to tonic periods during REM sleep without the FNE. Second, one study investigated 101 blood oxygenation level-dependent (BOLD) responses to acoustic stimulations (Wehrle 102 et al., 2007) and reported that acoustic stimulations during the tonic period induced 103 BOLD activations in the auditory cortex to some degree. In contrast, BOLD activations 104 in the auditory cortex during the phasic period were only minimal. Therefore, the ability 105 to process external stimuli may be reduced significantly during the phasic period but 106 sustained to some extent during the tonic period. Therefore, it is important to split REM 107 sleep into phasic and tonic periods for examination of evoked brain responses during 108

REM sleep. 109
The present study investigated whether the FNE affected the sleep depth and 110 vigilance of REM sleep, and whether decreased sleep depth or enhanced vigilance, if any, 111 shows interhemispheric asymmetry in an analogous manner to NREM sleep. Theta 112 activity was investigated as an index of the depth of REM sleep because it is one of the 113 major spontaneous brain oscillations during REM sleep (Takahara et al., 2006a). Our 114 previous study suggested that brain activities may need to be source-localized using 115 individual anatomical brain information for the detection of asymmetry in sleep depth 116 (Tamaki et al., 2016). Moreover, interhemispheric asymmetry in sleep depth is difficult to 117 identify on the sensor space, at least in the case of NREM sleep (Tamaki et al., 2016). 118 Therefore, we tested whether the FNE during REM sleep involved interhemispheric 119 asymmetry in theta activity in the DMN using a source-localization technique that 120 combines EEG, structural MRI, and PSG in the sleeping brain. To test whether the FNE 121 altered vigilance, we measured evoked brain responses to external stimuli using an 122 oddball paradigm during tonic and phasic periods. An evoked brain potential during 123 REM sleep, the latency of which is approximately 200 ms (hereafter, P2), is elicited to 124 deviant tones (Bastuji et al., 1995a;Sallinen et al., 1996;Takahara et al., 2002;2006b). 125 Previous studies found that the amplitude of P2 reflected the degree of vigilance during 126 REM sleep. Therefore, we used the P2 amplitude to test whether vigilance was enhanced 127 in one hemisphere with the FNE analogously to NREM sleep. We also tested whether the 128 impact of the FNE on brain responses differed between the phasic and tonic periods 129 because the FNE may differentially influence these periods. Sixteen subjects participated in the present study. Eight subjects participated in 135 Experiment 1 (6 females; 22.3 ± 0.84 yrs, mean ± SEM), and the other eight subjects 136 participated in Experiment 2 (4 females; 24.4 ± 0.86 yrs, mean ± SEM). An experimenter 137 thoroughly described the purpose and procedures of the experiment to candidate subjects, 138 and they were asked to complete questionnaires about their sleep-wake habits for 139 screening, including usual sleep and wake times, regularity of their sleep-wake habits and 140 lifestyle, habits of nap-taking, and information on their physical and psychiatric health, 141 including sleep complaints. The exclusion criteria included a physical or psychiatric 142 disease, currently receiving medical treatment, suspected sleep disorder, and habits of 143 consuming alcoholic beverages before sleep or smoking. Eligible people had regular 144 sleep-wake cycles, i.e., difference between average bedtimes and sleep durations on 145 weekdays and weekends was less than 2 h, and the average sleep duration ranged from 6 146 to 9 h regularly. All subjects gave written informed consent for their participation in 147 experiments. Data collection was performed at Brown University. The institutional 148 review board approved the research protocol. 149 Two subjects' data from day 1 and another 2 subjects' data from day 2 were 150 excluded from Experiment 2, resulting in the total number of data 6 each for days 1 and 151 2. The data omitted from day 1 and day 2 were different subjects. The reasons for data 152 omission were lack of REM sleep (n=2), lack of sound presentation due to arousals 153 during REM sleep (n=1), or the measured brain responses were too noisy (n=1). 154 155

Experimental design 156
Subjects in Experiments 1 and 2 were instructed to maintain their regular sleep-wake 157 habits before the experiments started, i.e., their daily wake/sleep time and sleep duration 158 until the study was over. The sleep-wake habits of the subjects were monitored using a 159 sleep log for 3 days prior to the experiment. Subjects were instructed to refrain from 160 alcohol consumption, unusual excessive physical exercise, and naps on the day before the 161 sleep session. Caffeine consumption was not allowed on the day of experiments.

Experiment 2 174
We measured the evoked brain responses in each hemisphere during REM sleep using an 175 oddball paradigm. We tested whether the evoked brain responses to deviant sounds were 176 larger on day 1 than day 2. Eight subjects participated in two experimental sleep sessions 177 (day 1 and day 2). These sessions were performed approximately one week apart so that 178 any effects of napping during the first sleep session would not carry over to the second 179 sleep session. traditional passive electrode systems. The data quality with active electrodes was as good 203 as 5 kΩ using passive electrodes, which were used for EOG and EMG (BrainAmp ExG, 204 Brain Products, LLC). Horizontal EOG was recorded from 2 electrodes placed at the 205 outer canthi of both eyes. Vertical EOG was measured from 2 electrodes 3 cm above and 206 below both eyes. EMG was recorded from the mentum (chin). ECG was recorded from 2 207 electrodes placed at the right clavicle and the left rib bone. The impedance was kept 208 below 10 kΩ for the passive electrodes. Brain Vision Recorder software (Brain Products, 209 LLC) was used for recording. The data were filtered between 0.1 and 40 Hz.

Classification of phasic and tonic periods 261
To classify the EEG recordings during REM sleep into phasic or tonic periods in 262

Analysis of brain responses to auditory stimuli 292
We examined the evoked brain potential known as the P2 (Takahara et al., 2002;2006b) 293 from EEG data recorded during REM sleep using an oddball paradigm in Experiment 2. 294 P2 is a positive brain potential that appears during REM sleep, and its amplitudes to rare 295 and salient stimuli increase (Takahara et al., 2002;2006b). Therefore, these responses are 296 used as an index of vigilance during REM sleep in humans. 297 To obtain the P2, 6 frontal channels (3 channels per hemisphere) were analyzed 298 (left: FC1, FC3, FC5; right: FC2, FC4, FC6). The frontal region was chosen for the 299 current analysis because these areas are near the DMN, where interhemispheric 300 asymmetry was observed with the FNE during deep NREM sleep (Tamaki et al., 2016). 301 All data were examined visually for each trial, and any trials that included arousals 302 (Bonnet, 1992;Iber et al., 2007) or motion artifacts were excluded from further analyses. 303 Analyses of brain responses followed a previous study (Tamaki et al., 2016). The  Averaged values were obtained for each of the phasic and tonic periods (see 315

Statistical analyses 318
The α level (type I error rate) of 0.05 was set for all statistical analyses. In Experiment 1, 319 a two-tailed paired t-test was performed for analyses of theta activity. In Experiment 2, a 320 3-way repeated measures ANOVA was used for the analysis of P2. In post hoc tests, t-321 tests were used with Bonferroni correction. 322 323

Evoked brain response 353
We obtained the P2 brain potential (Figure 2), which is one of the primary evoked 354 components during REM sleep (Takahara et al., 2002;2006b). The amplitude of P2 may 355 differ by eye movement state. Therefore, we analyzed the brain responses for the tonic 356 To test whether the brain was vigilant in a brain hemisphere during REM sleep 359 with the FNE, and whether the vigilance level differed by eye movement state, we 360 measured the mean amplitudes of the evoked brain responses for each hemisphere (left 361 vs. right), period (phasic vs. tonic), and sound (deviant vs. standard) (see Analysis of 362 brain responses to auditory stimuli in Methods for measurement of amplitude). We 363 found that P2 was elicited only during the tonic period, and not during the phasic period, 364 on day 2 (Figures 2C and D for the ground-averaged brain responses on day 2, which 365 replicates previous studies). However, a large P2 was elicited to deviant tones during 366 tonic and phasic periods on day 1 with the FNE (Figures 2A and B for the ground-367 averaged brain responses on day 1). However, P2 was elicited only during the tonic 368 period on day 2, and not during the phasic period (Figures 2C and D for the ground-369 averaged brain responses on day 2). 370 A 3-way ANOVA with within-subjects factors of Period (tonic, phasic) and 371 Hemisphere (left, right) and a between-subjects factor of Day (day 1, day 2) was 372 performed on the P2 amplitudes elicited by deviant tones (Figure 3). The Day factor was 373 a between-subject factor due to data omission (see Participants above). If there was 374 interhemispheric asymmetry in the brain response associated with the FNE, then a 375 Because a Period x Day interaction was significant, we performed post hoc analyses to 385 investigate the source of this interaction. Because no significant effect of Hemisphere 386 was found in the above ANOVA, data from the left and the right hemispheres were 387 pooled for subsequent analyses. Post hoc t-tests indicated a significant difference in the 388 amplitude between day 1 vs. day 2 in the phasic period ( Figure 3B; unpaired t-test, t (24) 389 = 4.31, p = 0.002, Bonferroni correction for the 4 comparisons) and between the phasic 390 and tonic periods on day 2 ( Figure 3B; paired t-test, t (11) = 3.30, p = 0.029). There was 391 no significant difference between days in the tonic period ( Figure 3B; unpaired t-test, t 392 (22) = 1.57, p = 0.131) or between the tonic vs. phasic periods on day 1 (Figure 3B; 393 paired t-test, t (11) = 2.29, p = 0.171). We performed one-sample t-tests on the 394 amplitudes for each of the periods and days to investigate whether P2 was elicited, being 395 larger than zero, on days 1 and 2 for the phasic and tonic periods. (Figure 3B). The 396 amplitude was significantly different from 0 during the tonic periods on day 1 (t (11) = 397 7.92, p < 0.001, Bonferroni correction for the following 4 comparisons) and day 2 (t (11) 398 = 4.34, p = 0.005), and during the phasic period on day 1 (t (11) = 5.29, p < 0.001), but 399 not on day 2 (t (11) = 2.30, p = 0.168). 400 401

Discussion 402
The present study found that the vigilance level was higher on day 1 than day 2, 403 specifically during the phasic period. However, interhemispheric asymmetry in evoked 404 brain responses was not found during REM sleep. Theta activity during REM sleep did 405 not show interhemispheric asymmetry. 406 Notably, although the brain responses to auditory stimuli did not show 407 interhemispheric asymmetry associated with the FNE, a larger response to rare stimuli 408 was found, specifically during the phasic period. Because the amplitude in an evoked 409 brain response to deviant stimuli correlates with the degree of vigilance (Nielsen- observed. These results demonstrate that a night-watch system using both hemispheres 413 exists during the phasic period of REM sleep. 414 We found that the amplitudes of brain responses were larger on day 1 than day 2, 415 specifically during the phasic period. Why did the phasic period, but not the tonic period, 416 show augmented evoked brain responses in association with the FNE? The brain is more 417 sensitive in the monitoring of external stimuli during the tonic period than the phasic 418 period during normal sleep without the FNE (Takahara et al., 2002). Therefore, the 419 capacity for information processing or attention to external stimuli may already be 420 limited (Kahneman, 1973;Marois and Ivanoff, 2005) during the tonic period. If the 421 attentional resource is spared for external monitoring during the tonic period, then when 422 there is a need to increase vigilance even further during sleeping in an unfamiliar 423 environment, such an increase would have to occur outside the tonic period, i.e., during 424 the phasic period. 425 The phasic period during normal REM sleep without the FNE is linked to 426 subjective mental activities (Berger and Oswald, 1962;Weinstein et al., 1988). Notably, a 427 previous study suggested that the sensitivity to external stimuli was lowered when there 428 is sleep-onset dreaming (Michida et al., 2005). Therefore, the reason that external 429 monitoring is impaired during the phasic period of normal REM sleep without the FNE 430 may be due to ongoing mental activities, including dreaming, which may interfere with 431 the monitoring of the external environments (Sallinen et al., 1996;Michida et al., 2005). 432 However, the large brain response was elicited during the phasic period during REM 433 sleep in association with the FNE. This result suggests that resources for internal mental 434 activities are deployed for the external monitoring during sleeping in an unfamiliar 435 environment when the FNE occurs. 436 We did not find a clear interhemispheric asymmetry in theta activity or the 437 amplitude of the evoked potentials between hemispheres during REM sleep in association 438 with the FNE. These results contrast our previous study (Tamaki et al., 2016) in which 439 we found interhemispheric asymmetry in regional slow-wave activity and vigilance 440 during deep NREM sleep. This difference suggests that a different mechanism than deep 441 NREM sleep applies to a night-watch system during REM sleep. The present results 442 showed that the vigilance to deviant sounds increased on day 1, which was associated 443 with the FNE. Therefore, a type of surveillance to the external world may exist during 444 REM sleep. It may be the case that the arousal threshold may be too high and costly 445 during deep NREM sleep to increase vigilance in both hemispheres, and only one 446 hemisphere may be used for surveillance. However, there are already resources for 447 dreaming and mental activity during REM sleep. These resources may be used for 448 surveillance in both hemispheres. This deployment of resources may occur without much 449 sacrifice in sleep depth during REM sleep. 450 In conclusion, REM sleep has a protective mechanism that may involve a 451 different mechanism than deep NREM sleep. A night-watch during REM sleep was 452 shown as increased vigilance in both hemispheres throughout REM sleep, specifically 453 during the phasic period. This REM-specific night-watch system may be realized by 454 deploying the resources available to internal activity for surveillance during sleeping in 455 an unfamiliar environment.   The amplitudes for each phasic and tonic period on days 1 and 2 (hemispheres averaged).