Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jul 1.
Published in final edited form as: J Genet Psychol. 2021 Apr 19;182(4):236–251. doi: 10.1080/00221325.2021.1905596

Pre-Sleep Arousal and Sleep in Early Childhood

Caroline P Hoyniak 1, Maureen M McQuillan 2, John E Bates 3, Angela D Staples 4, AJ Schwichtenberg 5, Sarah M Honaker 2
PMCID: PMC8684049  NIHMSID: NIHMS1760198  PMID: 33870880

Abstract

Research suggests that arousal during the transition to sleep – “pre-sleep arousal”— is associated with sleep disturbances. Although a robust literature has examined the role of pre-sleep arousal in conferring risk for sleep disturbances in adults, substantially less research has examined the developmental origins of pre-sleep arousal in early childhood. The current study examined pre-sleep arousal using parent-report and psychophysiological measures in a sample of preschoolers to explore the association between different measures of pre-sleep arousal, and to examine how nightly pre-sleep arousal is associated with sleep. Participants included 29 children assessed at 54 months of age. Pre-sleep arousal was measured using parent-reports of child arousal each night at bedtime and using a wearable device that took minute-by-minute recordings of heart rate, peripheral skin temperature, and electrodermal activity each night during the child’s bedtime routine. This yielded a dataset with 4,550 minutes of ambulatory recordings across an average of 3.52 nights per child (SD=1.84; range 1–8 nights). Sleep was estimated using actigraphy. Findings demonstrated an association between parent-reported and psychophysiological arousal, including heart rate, peripheral skin temperature, and skin conductance responses during the child’s bedtime routine. Both the parent-report and psychophysiological measures of pre-sleep arousal showed some associations with poorer sleep, with the most robust associations occurring between pre-sleep arousal and sleep onset latency. Behavioral and biological measures of hyperarousal at bedtime are associated with poorer sleep in young children. Findings provide early evidence of the utility of wearable devices for assessing individual differences in pre-sleep arousal in early childhood.

Keywords: pre-sleep arousal, sleep, early childhood, hyperarousal

Arousal during the transition to sleep – termed pre-sleep arousal— is associated with both subjective and objective sleep disturbances (Espie, 2002; Garde et al., 2012). Pre-sleep arousal has been theoretically divided into two domains: 1) cognitive arousal—defined by increases in mentation that occur during the pre-sleep period, including worries and racing thoughts that are difficult to control, and 2) physiological/somatic arousal—distressing physical sensations of parasympathetic arousal (e.g., racing heart, muscle tension, jittery nervousness; Nicassio, Mendlowitz, Fussell, & Petras, 1985). A bi-directional association between pre-sleep arousal and sleep has been proposed, whereby increased arousal during the pre-sleep period leads to sleep disturbances, and these sleep disturbances lead to increased worry/arousal the subsequent night (Garde et al., 2012). Hyperarousal, defined as either an elevated basal arousal or difficulties in down-regulating arousal across a number of domains (e.g., body temperature, heart rate, etc.) throughout the day and night, often characterizes poor sleepers (Bonnett & Arand, 2010; Espie, 2002), and is conceptualized as a key feature of insomnia.

A robust literature has demonstrated associations between higher pre-sleep arousal and sleep problems/insomnia in adults (Espie, 2002; Wicklow & Espie, 2000), but substantially less research has focused on pre-sleep arousal in children. Specific findings on pre-sleep arousal in children are described in detail below, but broadly, evidence suggests that pre-sleep arousal predicts insomnia symptoms in school-aged children (Gregory et al., 2008), and that higher physiological arousal measured during the day is associated with worse nighttime sleep in school-aged children (El-Sheikh & Buckhalt, 2005; El-Sheikh, Erath, & Bagely, 2013; Elmore-Staton, El-Sheikh, Vaughn, & Ariswalla, 2012). However, missing from this literature is an examination of pre-sleep arousal in early childhood, including investigations of the association between physiological indicators of pre-sleep arousal and sleep patterns in young children.

Importantly, some aspects of pre-sleep arousal in early childhood may differ from adult pre-sleep arousal. For example, young children are unlikely to experience some aspects of cognitive arousal, such as worry about the consequences of poor sleep on their daytime functioning. However, other aspects of pre-sleep arousal are likely to be present across the lifespan, such as restlessness, over-excitement, and fearful or racing thoughts. Parent-led bedtime routines – a set of predictable activities aimed at accomplishing important evening tasks (e.g., brushing teeth) and providing children with assistance in reducing vigilance and arousal (e.g., reading a story, snuggling) – are widely used in early childhood and are consistently associated with better sleep (Mindell & Williamson, 2018). Bedtime routines are an important window of time for understanding how young children successfully transition to sleep, as well as the factors that make this transition more difficult, such as heightened pre-sleep arousal. To better understand pre-sleep arousal in early childhood, the current study examined arousal during bedtime routines using parent-reports and psychophysiological measures in a community sample of preschoolers. We measured several aspects of pre-sleep arousal that spanned both the cognitive and physiological/somatic domains. We aimed to (1) assess relations between methodologically different measures of pre-sleep arousal, and (2) to examine how nightly pre-sleep arousal is associated with actigraphically-derived sleep.

Measuring pre-sleep arousal

Pre-sleep arousal has been measured using a number of techniques, most commonly via self-reports of arousal levels around the initiation of sleep. Such measures assess pre-sleep arousal either on a global scale, examining typical levels of pre-sleep arousal for the individual, or on a night-to-night basis. These measures typically include separable components for cognitive and physiological/somatic arousal and are most commonly completed by the individuals themselves. However, preschool-aged children may not be able to reliably report on either their cognitions or their physiological states, thus parent-report of their child’s observable pre-sleep arousal levels may be useful as a proxy measure.

Parent reports of cognitive and physiological/somatic pre-sleep arousal should be associated with psychophysiological measures of arousal, to the extent that they are measuring the same construct. In the current study, we focus on three specific indexes of psychophysiological hyperarousal – heart rate, temperature, and electrodermal activity— all three of which have been hypothesized to be disrupted both throughout the day, as well as during the pre-sleep period, in adults with sleep disturbances (Bonnet & Arand, 2010; Monroe, 1967; Riemann et al., 2010). In children, however, data on the association between psychophysiological hyperarousal and subjective pre-sleep arousal are lacking. Heart rate, which is governed by both sympathetic and parasympathetic processes, predictably decreases prior to sleep onset (Stein & Pu, 2012). Higher heart rates during the transition to sleep have been found to characterize poor adult sleepers (Farina et al., 2013; Nelson & Harvey, 2003; Stein & Pu, 2012). The predictable pattern of core body temperature fluctuation, in which core temperatures tend to be at their lowest at 9:00 p.m. and at their highest at 5:00 a.m., is thought to be governed by the body’s circadian clock (Crowley, 2013; Gilbert, van den Heuvel, Ferguson, & Dawson, 2004). The decrease in core body temperature that accompanies sleep onset is thought to be primarily facilitated by increases in peripheral skin temperature. Increases in peripheral skin temperature facilitate peripheral heat loss, which in turn facilitates decreases in core body temperature (Gilbert et al., 2004). Thermoregulatory fluctuations (i.e., decreases in core body temperature and increases in peripheral body temperature prior to sleep onset) are thought to play a causal role in promoting sleep regulation (Gilbert et al., 2004). Growing evidence suggests that disrupted thermoregulation may be associated with the hyperarousal profile that characterizes many adults with insomnia (Lack et al., 2008).

Of similar interest are measures of electrodermal activity (EDA). EDA is an index of the activity of eccrine sweat glands, governed by the sympathetic nervous system (Dawson, Schell, & Filion, 2000). Daily EDA patterns show predictable, circadian-based fluctuations, and are at a minimum in the morning and at a maximum in the evening (Hot, Naveteur, Leconte, & Sequeira, 1999). EDA is divided into phasic and tonic components. The phasic components –skin conductance responses and non-specific skin conductance responses— are sensitive to episodic shifts in arousal, while the tonic components—skin conductance levels–reflect resting activity (Dawson et al., 2000). During wakefulness, skin conductance responses are thought to reflect attention, arousal, anxiety, and other emotional states (Critchley, 2002; Nikula, 1991). Self-reports of pre-sleep emotional states, including negative emotions, stress, and worry, are associated with an EDA profile of increased sympathetic activity throughout the subsequent sleep cycle (Delannoy et al., 2015; Lester et al., 1967). To our knowledge, very little research in either adults or children has examined EDA prior to sleep onset. EDA may serve as a psychophysiological index of pre-sleep arousal. Because EDA shows specific, circadian-based fluctuations, reaching a maximum in the evening and a minimum in the morning, individual differences in these circadian-based fluctuations may theoretically reflect the dysregulated hyperarousal that is characteristic of poor sleepers in general. However, to our knowledge, this has yet to be examined in preschool-aged children.

Pre-sleep arousal and sleep in childhood

A relatively sparse literature suggests an association between the cognitive and somatic symptoms associated with pre-sleep arousal and sleep disturbances in children. Higher parent and/or child-reported pre-sleep arousal has been found to be associated with increased sleep problems (Alfano et al., 2010; Bagely et al., 2015; Fisher et al., 1994; Gregory et al., 2008). Additionally, in school-aged children, individual reports of pre-sleep arousal have been associated with disrupted sleep as measured using both actigraphy (duration of night wakings; Bagely et al., 2015) and polysomnography (a lower percentage of REM sleep; Patriquin et al., 2014).

Several additional studies have considered how measures of trait-like psychophysiological arousal/regulation during the day are associated with nighttime sleep. In both school-aged children and preschoolers, individual differences in daytime physiological arousal have been shown to be associated with sleep problems (Alkon, Boyce, Neilands, & Eskenazi, 2017; Elmore-Staton et al., 2012; El-Sheikh et al., 2013; El-Sheikh & Arsiwalla, 2010; El-Sheikh & Buckhalt, 2005). In preschoolers, Elmore-Staton et al. (2012) reported that higher resting state respiratory sinus arrhythmia, a marker of parasympathetic nervous system activation (an index of physiological regulation), was associated with less sleep activity and higher sleep efficiency. In contrast, in preschoolers, Alkon et al. (2017) did not find a direct association between either parasympathetic or sympathetic nervous system reactivity during a challenging task and sleep. However, studies examining the association between psychophysiological arousal and sleep in children measure psychophysiological arousal during the day, and it is unclear how this would generalize to psychophysiological processes closer to bedtime. When considering temperature, Okamoto-Mizuno, Mizuno, and Shirakawa (2016) examined the association between core and peripheral skin temperatures and sleep in 18 preschoolers. They reported higher distal/peripheral skin temperature was associated with actigraphally-derived sleep, including higher sleep efficiency, fewer and shorter wake episodes after sleep onset, and less activity during sleep. However, in Okamoto-Mizuno et al. (2016) skin temperature was only measured during the sleep cycle, so it is unclear how changes in skin temperature specifically during the pre-sleep period may be associated with sleep.

To our knowledge, no research has explored how pre-sleep arousal, measured using either parent-reports or psychophysiological data, is associated with sleep in early childhood. Sleep problems are thought to emerge early in life and remain relatively stable across development (Gregory & O’Connor, 2002; Williams et al., 2017). Additional knowledge about pre-sleep arousal in early childhood, alternative ways to assess it, and its associations with sleep parameters will elucidate possible contributions to sleep disturbances in early childhood. Additionally, researchers and clinicians have long speculated that childhood behavioral sleep interventions improve sleep outcomes by decreasing arousal levels at bedtime (Mindell, Telofski, Wiegand, & Kurtz, 2009), but the association between arousal levels at bedtime and sleep has not yet been established in research.

Current study

The current study examined: 1) the correspondence between parent-report and psychophysiological measures of pre-sleep arousal, and 2) how both measures of pre-sleep arousal were associated with actigraphic measures of sleep in a community sample of preschoolers. Pre-sleep arousal was quantified in two ways: 1) with nightly parent-reports of child arousal at bedtime, and 2) with ambulatory measures of child heart rate, temperature, and EDA during the child’s bedtime routine. Our measures of pre-sleep arousal spanned both the cognitive and physiological/somatic domains, with our measure of parent-reported pre-sleep arousal including items that assessed aspects of both theoretical domains of pre-sleep arousal and our psychophysiological measures assessing the physiological/somatic domain. We hypothesized that there would be an association between the different measurement modalities of pre-sleep arousal. In particular, we expected that higher levels of parent-reported arousal would be associated with faster heart rates, lower peripheral skin temperatures (which may be a proxy index of higher core body temperatures), and altered EDA. Because EDA typically shows a cyclical pattern, reaching a maximum in the evening, we expected that children whose parents described them as showing high levels of pre-sleep arousal would also show more deviation from the traditional circadian-based, cyclical EDA pattern by having lower skin conductance levels and fewer non-specific skin conductance responses during the pre-sleep period compared to their peers. Additionally, we expected that higher levels of pre-sleep arousal (both parent report and psychophysiological) would be associated with shorter sleep durations, later sleep schedules, increased sleep activity, and longer sleep onset latencies.

Methods

Participants

Participants included 36 children assessed at approximately 54 months of age. Participants were recruited from a wider longitudinal study of toddler development that measured children at 30, 36, 42, and 54 months [Hoyniak et al., 2019]. Participants were initially recruited at 30 months using: (1) a database search based on county birth records, (2) community partnerships (e.g., Head Start and the Housing Authority), and (3) public advertisements (e.g., postcards, flyers, bus advertisements). All children scheduled for a 54-month assessment between February 2018 to May 2019 were invited to wear the Empatica E4 multi-sensor device (henceforth referred to as an “E4”; n = 40), with an acceptance rate of 90%. The current study focused on this subsample of children (n = 36) who agreed to attempt to complete additional, exploratory measures of pre-sleep arousal (including wearing the E4 device) during their 54-month assessment. Usable E4 data were collected from 29 children (14 females). Non-usable E4 data were due to child refusal to wear the device, technical malfunctions, and in one case, a dog chewing on the device. The final sample of 29 children had a mean age of 54.74 months at the time of actigraphy data collection (SD = 0.63 months; range: 53.87 – 56.26. months) and was predominantly Caucasian (90%, 3% Latino, and 7% Other), from two-parent households (93%, 7% Single Parent), with a college-educated primary caregiver (76% college degree, 13% some college, 4% high school diploma or less, 7% unknown or not reported). Family socioeconomic status (SES), calculated using the Hollingshead Four Factor Index (Hollingshead, 1975), ranged from 16 to 66, with M = 51.09 (SD = 12.19), suggesting that the sample was predominantly middle class.

Procedures

For one week, children wore an actigraph (for as much of each 24-hour period as the child would tolerate) and an E4 device (during the hour or so leading up to bedtime each night). Simultaneously, primary caregivers were asked to complete nightly child sleep diaries that included questions about their child’s sleep patterns and parental perception of the child’s pre-sleep arousal. Informed consent was acquired. Compensation included $30 for the parents (combined payment for a home visit along with actigraph/sleep diary participation) and small wrapped toys worth approximately $1 each for the child to be given by parents in the morning for each night the child successfully wore the actigraph and E4 device. All procedures were approved by the Institutional Review Board at Indiana University.

Measures

Parent-reported pre-sleep arousal: Sleep diaries.

Primary caregivers (97% mothers) completed daily child sleep diaries, recording information about their child’s nightly bedtime routine and sleep pattern. Sleep diaries were completed concurrently with actigraph data collection to facilitate actigraphy scoring. Sleep diaries were completed for one to two weeks, with an average of 7.48 nights (SD = 1.64 nights). Each night, primary caregivers also reported on their child’s pre-sleep arousal by rating how “calm”, “wired”, “restless”, and “worried” their child seemed during their bedtime routine. The items assessing how “calm”, “wired”, and “restless” the child was during their bedtime routine theoretically mapped onto the physiological domain of pre-sleep arousal, while the “worried” item more closely mapped onto the cognitive domain of pre-sleep arousal. Each item was rated on a scale from 1 (not at all) to 5 (extremely), with higher scores indexing higher arousal at bedtime. The four items were considered individually and summed to create a composite, for each night. For the arousal composite, the “calm” item was reverse scored prior to inclusion. The Cronbach’s alpha value for this composite index was 0.72. A sample night of the child sleep diary, including the pre-sleep arousal questions, is provided in Supplemental Appendix A.

Psychophysiological measures of pre-sleep arousal: Empatica E4 multi-sensor.

Child pre-sleep arousal indexes of heart rate, peripheral skin temperature, and EDA were recorded using an E4 device. All E4 data were analyzed using a series of scripts (summarized below) in MATLAB, which were developed as a part of another research project (Schwichtenberg & O’Haire, 2017). Children wore the E4 device during the hour or so leading up to their bedtime, and parents were given the option to remove the device when the child went to sleep, during the night while the child was asleep, or to have their child wear the device throughout the night. In the current study, children wore the E4 device on their dominant wrist, as the actigraphs were worn on the non-dominant wrist. EDA was analyzed by splitting the collection period into one-minute intervals. In the current study, skin conductance responses were not examined in an event-related context, so all skin conductance responses were considered to be non-specific. For each minute, measured phasic elements included the number of non-specific skin conductance responses, the peak non-specific skin conductance response amplitude in that minute, and the average non-specific skin conductance response amplitude across the minute. Minute-by-minute tonic levels (i.e., skin conductance levels) reflected skin conductance levels in the absence of phasic non-specific skin conductance responses. Each epoch of processed data was examined for usability following the guidelines provided by Kleckner and colleagues (2018). A data point was excluded if it failed to meet any of the following criteria: (1) EDA was within the range of 0.5 and 60 microsiemens (μS); (2) change in EDA per second within 10 μS; (3) heart rate within the range of 60 – 120 beats per minute; and (4) temperature within the range of 30 to 40 degrees Celsius. For each invalid second, the 5 seconds before and after were also excluded (Kleckner et al., 2018). Within each minute, at least 42 seconds (70% of that minute) had to be retained based on the criteria above to be included in analyses. The use of this cutoff was a functional one reflecting our desire to maximize data use while maintaining high levels of data integrity.

Data analysis was restricted to minutes that occurred during the child’s bedtime routine. The timing of the child’s bedtime routine was determined by parent reports on the daily child sleep diaries, which indicated when the child’s bedtime routine began and ended each night. This resulted in a final usable sample of 4550 minutes across 29 participants (16.89% of the data were excluded based on the criteria above). Averages of each EDA index, both tonic (i.e., skin conductance level) and phasic (i.e., non-specific skin conductance responses, peak non-specific skin conductance response amplitude, and average non-specific skin conductance response amplitude), peripheral skin temperature, and heart rate were calculated across the usable intervals of the child’s bedtime routine.

Child sleep: Actigraphy.

Child sleep was assessed using MicroMini Motionlogger actigraphs from Ambulatory Monitoring, Inc. (Ardsley, NY), watch-like accelerometers worn on the non-dominant wrist that monitor minute-by-minute motor activity to determine sleep and wake patterns. Actigraphy data were processed using the Sadeh algorithm (validated for use with children; Sadeh, Sharkey, & Carskadon, 1994) by research assistants who were trained to reliability with one another. For the current study, we derived three sleep composites for each night based on a Principal Components Analysis (PCA) of the actigraphy variables exported from AW2 software (Staples, Bates, Petersen, McQuillan, & Hoyniak, 2019). Composites were used instead of single, raw actigraphy variables to get more robust measures of the sleep constructs of interest and to reduce the number of variables tested in relation to pre-sleep arousal. The three composites, sleep duration, sleep timing, and sleep activity, represent broad dimensions of sleep that are often examined in the child sleep literature (Meltzer, Montgomery-Downs, Insana, & Walsh, 2012). Table 1 contains descriptive statistics for the various actigraphy variables included in each composite. Each composite showed adequate internal reliability (sleep duration α = 0.81, sleep timing α = 0.95, and sleep activity α = 0.72). These composites were used in all models, along with a single index of sleep onset latency, a variable that did not load onto any of the sleep composites yet is clinically relevant in the assessment of childhood sleep difficulties.

Table 1.

Actigraphy variables included in each composite, along with means and standard deviations

Composite Variable Names M(SD)
Sleep Duration Avg. sleep period (time from sleep onset to waketime; min) 597.30 (56.52)
Avg duration of time in bed (min) 638.27 (51.33)
Avg. min asleep in bed (minutes scored as sleep during the time in bed) 533.62 (78.49)
Sleep Timing Avg. time of midsleep (HH:MM in 24-hour time) 01:42 (59 min)
Avg. time of sleep onset (HH:MM in 24-hour time) 20:44 (1 hr, 13 min)
Avg. bedtime (HH:MM in 24-hour time) 20:08 (1 hr, 13 min)
Sleep Activity Avg. time (min) awake after sleep onset 62.88 (45.52)
SD of avg. min to min activity levels 30.11 (10.75)
Avg number of awakenings (lasting 5 min or more) 3.74 (3.39)
Avg. duration (min) of longest wake episode (after sleep onset) 16.23 (14.15)
Avg. percent of active epochs (after sleep onset) 39.01 (13.42)
Open in a new tab

Analysis Plan

Analysis only focused on nights when children had useable E4 data, which ranged from 1 to 8 nights with a mean of 3.52 nights (SD = 1.84) per child. This range of nights reflects that some families had their child wear the E4 device for longer periods of time (e.g., all night), and thus the battery on the E4 device ran down after only a few days. Initially, families were given instructions to charge their E4 device when it ran down, but this substantially increased perceived subject burden. We then adjusted the protocol, instructing caregivers to have their child wear the E4 device as many evenings of the week as possible until the battery ran down. Of the nights with usable E4 data, 4% of those nights were missing parent-reported pre-sleep arousal and 26% were missing actigraphy data (primarily due to actigraph failure). Missingness was handled using pairwise deletion.

Analysis proceeded in two steps. We first examined the correspondence between parent report and psychophysiological measures of pre-sleep arousal each night. We initially used Pearson correlations, but most children contributed more than one night of data so the assumption of independence inherent in Pearson correlations was violated. Given the nested nature of the data (nights nested within children), we re-examined significant correlations using multiple regression with a cluster variable (i.e. clustered/nested regression), a statistical technique that accounts for dependency in data caused by single individuals contributing multiple measurement occasions. We still present the Pearson correlation coefficients, though, because these values are more easily understandable. However, we only interpret the associations that remained significant when using the nested regression approach. Nested regression models were fit using the rms package (Harrell Jr, 2019) in R (R Core Development Team, 2014), which calculates robust standard errors using a robust (Huber-White sandwich) estimator of the covariance matrix (Huber, 1967; White, 1980). Sandwich estimators are widely used to account for data dependency in regression models (Wang, 2014). Next, we examined the nightly associations between both parent report and psychophysiological measures of pre-sleep arousal and sleep. We used the same analytic approach to examine the association between pre-sleep arousal and sleep measures, first using Pearson correlations and then using nested regression. In the nested regression analysis, we also controlled for child sex.

Results

Descriptive statistics for variables included in the analyses are provided in Table 2.

Table 2.

Descriptive statistics for variables included in analysis

M SD Range
Parent-Reported Pre-Sleep Arousal Calm 2.84 1.36 1 - 5
Wired 1.69 1.06 1 - 5
Restless 2.11 1.19 1 - 5
Worried 1.35 0.76 1 - 4
Arousal Composite 8.20 3.31 4 - 16
Psychophysiological Pre-Sleep Arousal Mean Number NSSCRs 3.91 1.78 0 - 7
Mean Largest NSSCR (μS) 0.02 0.02 0 – 0.12
Mean Amp NSSCR (μS) 0.86 0.83 0 - 4.04
Mean SCL (μS) 7.79 10.33 0.06 - 49.67
Mean Temperature (Celsius) 34.70 1.36 30.80 - 40.64
Mean Heart Rate (beats per minute) 88.49 9.24 67.87 - 108.81
Child Sleep Sleep Duration 0.01 0.92 −3.08 – 2.11
Sleep Timing 0.00 0.97 −3.48 - 3.47
Sleep Activity 0.00 0.84 −1.35 – 1.90
Sleep Onset Latency (min) 36.64 30.61 4 - 121
Open in a new tab

Note: NSSCR = non-specific skin conductance response; μS = microsiemens; SCL = skin conductance level

Association between Parent-reported and Psychophysiological Pre-Sleep Arousal

A summary of all analyses completed to address this aim are provided in Tables 3 and 5. Higher heart rates during the child’s bedtime routine were associated with parent-reports of children being less calm, more restless and more aroused overall during the bedtime routine. Additionally, lower peripheral temperatures (which may index higher core temperatures) during the child’s bedtime routine were associated with parent-reports of children being less calm, more restless, and more aroused overall during the bedtime routine. Of the electrodermal indexes, only the average number of non-specific skin conductance responses per minute during the bedtime routine was associated with parent-reported restlessness, such that a lower mean number of non-specific skin conductance responses per minute was associated with more parent-reported restlessness. Controlling for child sex in these analyses did not change the pattern of significant findings.

Table 3.

Pearson correlations between parent and psychophysiological measures of pre-sleep arousal

Calm Wired Restless Worried Arousal Composite
Mean Number
NSSCRs/minute .12 .01 −.23* .00 −.10
Mean Largest NSSCR .09 .02 −.14 −.07 −.06
Mean Amp NSSCR .13 .00 −.16 −.08 −.10
Mean SCL .05 −.02 −.17 −.14 −.10
Mean Temperature .31** −.10 −.34** −.04 −.29**
Mean Heart Rate −.38** .26** .43** .11 .39**
Open in a new tab

Note:

*

p ≤ .05

**

p ≤ .01 – reflect significant of initial Pearson correlations, all two-tailed

Gray boxes indicate that the correlation remained significant at the p < .05 level when accounting for potential dependency in the data caused by some individuals contributing more than one night of data. Parent report variables are listed in columns and E4 variables in rows.

NSSCR = non-specific skin conductance response; SCL = skin conductance level

Table 5.

Association between parent and psychophysiological measures of pre-sleep arousal using nested regression

B SE p-value
Restless
  Mean Number −.16 .05 .003
  NSSCRs/minute
  Female −.13 .36 .72
  Mean Temperature −.30 .11 .01
  Female −.09 .34 .79
  Mean Heart Rate .06 .01 < .001
  Female −.14 .30 .63
Calm
  Mean Temperature −.32 .12 .008
  Female −.43 .41 .29
  Mean Heart Rate .06 .02 .001
  Female −.47 .34 .17
Wired
  Mean Heart Rate .03 .02 .08
  Female .41 .30 .17
Arousal Composite
  Mean Temperature −.68 .25 .008
  Female .23 1.02 .82
  Mean Heart Rate .14 .04 .001
  Female .08 .84 .93
Open in a new tab

Note: NSSCR = non-specific skin conductance response

Association between Parent-reported Pre-Sleep Arousal and Sleep

A summary of all analyses completed to address this aim is provided in Tables 4 and 6. Findings generally indicated that higher levels of parent-reported pre-sleep arousal were associated with worse sleep. Children who were rated as being less calm at bedtime had a later sleep schedule on that night (i.e., later bedtime, later mid-sleep, and later wake time). Children who were reported to show more wired behaviors at bedtime had, surprisingly, less active sleep that night. Additionally, children who were reported to be less calm, more restless, and more aroused overall during their bedtime routine took longer to fall asleep that night (sleep onset latencies were 8.56 minutes longer for each one-point decrease in calm arousal scale and 10.71 minutes longer for each one-point increase in the restlessness scale). Controlling for child sex in these analyses did not change the pattern of significant findings.

Table 4.

Pearson correlations between measures of pre-sleep arousal and sleep

Sleep Duration Sleep Timing Sleep Activity Sleep Onset Latency
Parent-Reported Pre-Sleep Arousal Calm .28* −.27* −.10 −.38**
Wired .17 .13 −.26* .04
Restless −.13 .27* .14 .43**
Worried −.06 .08 −.04 −.06
Arousal Composite −.10 .21^ .00 .35**
Psychophysiological Pre-Sleep Arousal Mean Number
NSSCRs/minute .34** −.20^ −.09 −.07
Mean Largest
NSSCR .16 −.15 −.01 .05
Mean Amp NSSCR .18 −.17 −.01 .02
Mean SCL .23^ −.09 −.05 −.29*
Mean Temperature .22^ −.13 −.27* −.24*
Mean Heart Rate −.17 .20^ .10 .51**
Open in a new tab

Note:

^

p ≤ .10

*

p ≤ .05

**

p ≤ .01 – reflect significant of initial Pearson correlations, all two-tailed

Gray boxes indicate that the correlation remained significant at the p < .05 level when accounting for potential dependency in the data caused by some individuals contributing more than one night of data

NSSCR = non-specific skin conductance response; SCL = skin conductance level

Table 6.

Association between parent and psychophysiological measures of pre-sleep arousal and sleep using nested regression

B SE p-value
Sleep Duration
  Calm −.19 .10 .07
  Female .01 .30 .98
  Mean Number .19 .07 .01
  NSSCRs/minute
  Female .33 .26 .22
Sleep Timing
  Calm .17 .09 .08
  Female −.21 .41 .61
  Restless .21 .12 .07
  Female −.27 .36 .46
  Arousal Composite .06 .04 .17
  Female −.29 .37 .44
Sleep Activity
  Wired −.22 .07 .003
  Female .12 .31 .70
  Mean Temperature −.17 .12 .15
  Female .03 .31 .93
Sleep Onset Latency
  Calm −8.71 3.97 .03
  Female 1.34 9.18 .88
  Restless 10.63 4.44 .02
  Female −7.13 9.47 .45
  Arousal Composite 3.21 1.25 .01
  Female −6.01 10.52 .57
  Mean SCL −1.18 .48 .01
  Female − 13.43 11.60 .25
  Mean Temperature −5.51 4.64 .24
  Female −7.31 11.20 .52
  Mean Heart Rate 1.66 .57 .005
  Female −7.70 9.64 .43
Open in a new tab

Note: NSSCR = non-specific skin conductance response; SCL = skin conductance level

Association between Psychophysiological Pre-Sleep Arousal and Sleep

A summary of all analyses completed to address this aim is provided Tables 4 and 6. There was a significant association between sleep duration and mean number of non-specific skin conductance responses per minute. Children with higher numbers of non-specific skin conductance responses per minute during their bedtime routine had longer sleep durations. Additionally, there was a significant association between sleep onset latency and both skin conductance levels (the tonic component of EDA) and heart rate. Children with lower skin conductance levels and higher heart rates during their bedtime routine took longer to fall asleep. Controlling for child sex in these analyses did not change the pattern of significant findings.

Discussion

The current study examined the association between parent-reported and psychophysiological measures of pre-sleep arousal in a community sample of preschoolers, as well as how pre-sleep arousal is associated with actigraphic sleep. Our hypotheses were partially supported; there was a significant association between parent-reported pre-sleep arousal and several indexes of psychophysiological arousal during the child’s bedtime routine. Specifically, a higher average heart rate, a lower average peripheral skin temperature, and a lower average number of skin conductance responses per minute, were all associated with higher levels of parent-reported arousal at bedtime. Both parent-reports and psychophysiological measures of pre-sleep arousal were associated with poorer actigraphic sleep, with the most robust associations emerging between pre-sleep arousal and sleep onset latency. Contrary to our hypotheses, there were few associations between either parent-reported or physiological pre-sleep arousal and measures of sleep duration, sleep activity, and sleep timing. Despite this, these findings provide the first evidence, to our knowledge, that pre-sleep arousal, measured both by parent reports and physiological data, is associated with sleep in early childhood.

The findings of the current study indicate that a number of psychophysiological measures may be useful for indexing the physiological/somatic aspects of pre-sleep arousal in young children, including elevated heart rates, lower peripheral skin temperatures (which may index higher core temperatures), and a lower number of non-specific skin conductance responses per minute. All three of these measures (i.e., heart rate, peripheral skin temperature, and EDA) are regulated by the autonomic nervous system and have been shown to be affected by sympathetic activation and parasympathetic deactivation surrounding arousal. Additionally, both skin temperature and EDA are thought to be regulated by circadian processes. Therefore, disruptions in these indexes may be associated, mechanistically, with disrupted sleep. Core temperatures are thought to be at their lowest in the evenings, while peripheral skin temperatures are at their highest in the evenings. Individual differences in both core and peripheral temperatures (i.e., higher core temperatures and lower peripheral temperatures) might reflect disruptions in circadian processes corresponding with hyperarousal that is associated with disturbances in sleep-wake regulation. Similarly, skin conductance levels are relatively higher in the evenings (Hot et al., 1999), and individual differences in these values (i.e., lower skin conductance levels in the evenings) might be associated with disruptions in circadian process. Disturbances in these measures (e.g., higher core temperatures in the evenings) have been found to be characteristic of adults with insomnia (e.g., Bonnet & Arand, 2010; Riemann et al., 2010), suggesting these indexes of hyperarousal are plausible biomarkers of insomnia (Fernandez-Mendoza, 2017). Our findings suggest that individual differences in these biomarkers, which perhaps both reflect and confer risk for sleep disruptions, may be present even in young children.

Importantly, our findings with the various measures of EDA (i.e., non-specific skin conductance responses and skin conductance levels) were not as consistent as our findings for heart rate and skin temperature. This could mean that EDA may be a weaker measure of pre-sleep arousal or sleep disruptions in young children. However, this could also be due to methodological limitations, including the fact that we used dry sensors to detect EDA. Future studies should examine how different technologies for assessing EDA may be useful for measuring pre-sleep arousal. The current study found an encouraging degree of correspondence between the parent-report and psychophysiological measures of pre-sleep arousal. These findings provide preliminary validation for both our short set of parent ratings and the psychophysiological measures of arousal. In their ratings, parents may have tuned into their children’s behavioral markers of physiological hyperarousal. Interestingly, not every item in the parent-report pre-sleep arousal measure showed an association with the physiological measures. The items indexing calmness and restlessness showed a larger number of significant associations with the physiological measures of arousal and sleep, while items indexing how wired and worried the child is appeared to be less useful. While we did find an association between parent reports of how “wired” the child was and mean heart rate, this finding did not remain significant when accounting for the longitudinal dependency in the data. It is possible that wired/tense behaviors in young children might be indicative of other problematic behaviors at bedtime (e.g., bedtime resistance, irritability) that do not truly reflect pre-sleep arousal and thus show limited correspondence with the physiological measures of pre-sleep arousal. Additionally, the fact that no association emerged between parent-reported worry at bedtime (i.e., “My child seems worried or has difficulty settling his/her mind”) and either physiological pre-sleep arousal or actigraphic sleep contrasts with the literature suggesting that cognitive pre-sleep arousal, evidenced by the increased presence of worries/racing thoughts during the pre-sleep period, is associated with disturbed sleep (Morin, Rodrigue, & Ivers, 2003). Our lack of findings in this domain may reflect the nuances of assessing worry in young children. Young children have more difficulty verbalizing internal emotional states and cognitions, and parents may fail to notice symptoms such as worry because of their often subtle presentations in early childhood (Whalen, Sylvester, & Luby, 2017). It is possible that parents of young children are better able to identify pre-sleep arousal using specific behavioral cues such as “calmness” and “restlessness”, rather than more vague cues such as “worried’ or “tense”. While we are not able to explore these possibilities in our current sample, future research focused on the construct of pre-sleep arousal in early childhood may help better elucidate the nuances of assessing pre-sleep arousal in very young children and the strengths/limitations of parent-reports of this arousal.

Although our measure of parent-reported pre-sleep arousal was short, and by no means comprehensive, items measuring pre-sleep arousal are not present in many of the scales commonly used to assess sleep in children. The findings that higher levels of parent-reported pre-sleep arousal were associated with both psychophysiological makers of hyperarousal as well as poorer sleep suggest that it might be useful to include additional items assessing child pre-sleep arousal in sleep assessment scales. Additional research on assessing parental reports of pre-sleep arousal in young children will be a crucial next step. Specifically, there is a need for identifying ways to more adequately assess the cognitive aspects of pre-sleep arousal in early childhood, given poor reporting capacities in young children and parent difficulties in identifying internal child emotional/cognitive states.

The findings of the current study also extend a robust literature on the role of hyperarousal in conferring risk for sleep disturbances in adults/youth to the early childhood period, a developmental era when sleep schedules are undergoing normative changes and sleep problems are likely to emerge (Galland, Taylor, Elder, & Herbison, 2012; Gregory & O’Connor, 2002; Williams et al., 2017). Dahl (1996) proposed that sleep and vigilance are opponent processes, and the initiation of sleep requires the down-regulation of arousal. The current study provides evidence for this, indicating that both behavioral, including parental perceptions of calmness and restlessness, and physiological measures of hyperarousal during the pre-sleep period, including number of non-specific skin conductance responses, skin conductance levels, and heart rate, are associated with subsequently poorer sleep in young children. Although several domains of sleep were shown to be worse in children with higher levels of pre-sleep arousal, including sleep duration, timing, and activity, the most robust associations emerged between pre-sleep arousal and sleep onset latency. Psychophysiological pre-sleep arousal is more clearly associated with sleep onset latency than with sleep throughout the night. This corresponds with a literature suggesting that pre-sleep arousal is associated with longer sleep onset latency in adults (Wicklow & Espie, 2000; Wuyts et al., 2012). In contrast to findings with adults, peripheral skin temperature during the bedtime routine was not associated with worsened sleep in nested regression models, despite evidence in our sample that peripheral skin temperature was associated with parent reports of pre-sleep arousal, and that there were modest Pearson correlations between peripheral skin temperatures and sleep activity and sleep onset latency. The lack of findings in this domain could be due to our relatively small sample size due to the exploratory nature of the current study. Conversely, our pattern of findings may reflect a developmental difference in the association between peripheral skin temperature and sleep. It could also reflect processes operating in only some, but not all children. Future studies with a larger sample size will be crucial for fully exploring the association between peripheral skin temperature and sleep in children.

A common feature of pediatric behavioral sleep interventions involves encouraging parents to implement consistent bedtime routines and to facilitate the down regulation of vigilance in their children during these routines, such as by including soothing activities like bathing, reading, and cuddling. Clinicians who introduce such interventions may consider assessing the cognitive and physiological/somatic aspects of child pre-sleep arousal before, during, and after the interventions to determine if these activities effectively reduce children’s pre-sleep arousal and promote an easier transition to sleep. Additionally, given the relative accessibility of wearable technology for monitoring heart rate and skin temperature, E4-type devices may be a practical addition to clinical assessment of the physiological/somatic aspects of pre-sleep arousal, especially when working with non-verbal patients or patients who are displaying more subtle, difficult to discern, symptoms of hyperarousal near bedtime. The current study provides evidence of the utility of measures of heart rate and peripheral skin temperature (and to some extent, EDA) in assessing pre-sleep arousal and propensity towards sleep difficulties in young children. However, additional replications, with larger samples of children as well as clinical samples of children with identified sleep disorders, will be crucial for further demonstrating the clinical utility of psychophysiological measures of pre-sleep arousal.

Strengths and Limitations

The current study has a number of important strengths. It included comprehensive measures across multiple days of sleep and pre-sleep arousal, using both psychophysiological and parent-reported behavioral measures. Many measures of pre-sleep arousal have parents and/or children report globally on their arousal levels at bedtime – generating a single index of pre-sleep arousal for an individual. While this approach is useful for individuals who have persistent difficulties with arousal at bedtime and can provide global estimates of the impact of pre-sleep arousal on sleep/functioning, there is likely a substantial amount of individual variability in arousal levels across time. The current study, which assessed pre-sleep arousal nightly, allowed the focus of analysis to be on night-by-night associations between arousal and sleep. This nightly approach may be especially useful for children as their behavior tends to be less crystalized across time and situations compared to adults, and children’s arousal levels in particular may vary substantially from night-to-night depending on family factors (e.g., family conflict and chaos) and next-day demands (e.g., school).

There are also limitations worth noting. Our sample was relatively small, primarily white, and middle-income. Given an established robust association between SES and sleep in early childhood (Williamson & Mindell, 2020), it will be important to replicate these results across larger and more generalizable samples, with a range of incomes (Schwichtenberg et al., 2019). Additionally, the current study focused on a community sample of children, and as such, the children included in this sample had relatively low levels of sleep problems. It is unclear how the association between pre-sleep arousal and sleep might vary in young children already displaying clinical sleep disturbances. Finally, the final sample in this study included only 29 children, and a larger sample would be useful. However, the sample size is comparable with recently published research on this topic (Okamoto-Mizuno et al., 2016), and these children provided dense data, including 4,550 minutes of E4 recordings and over 100 nights of assessment across subjects. This allowed us, despite the relatively small sample size, to answer important questions about the measurement of pre-sleep arousal and its association with sleep, a novel contribution to this field.

Conclusions

Findings of the current study demonstrate an association between parent-reports and psychophysiological measures of pre-sleep arousal in young children and suggests that pre-sleep arousal might be associated with poorer sleep in early childhood. These findings indicate that behavioral and biological measures of hyperarousal at bedtime are associated with poorer sleep in young children, especially a longer sleep onset latency. Our findings have important implications for subjective and physiological quantification of pre-sleep arousal in both research and clinical practice, providing early evidence of the utility of both a few parent-report items as well as wearable devices for assessing individual differences in pre-sleep arousal in early childhood. Additionally, our findings provide important information about a possible process, physiological/somatic pre-sleep arousal, that may contribute to sleep disturbances in early childhood.

Supplementary Material

Supp 1

Acknowledgments:

This project was funded with support from the Indiana Clinical and Translational Sciences Institute (National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award; UL1TR001108) and funds from Indiana University. CPH was supported by the National Institute of Mental Health training grant (NIMH T32 MH100019–06; PIs: Barch & Luby). AJS was supported by the National Institute of Mental Health (R00 MH092431) and the Gadomski Foundation. SMH was supported by National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences Award (UL1TR001108; PI: Shekhar).

Footnotes

The authors have nothing to disclose.

Data Availability Statement:

The data that support the findings of this study are available from the corresponding author, C.H., upon reasonable request.

References

  1. Alfano CA, Pina AA, Zerr AA, & Villalta IK (2010). Pre-sleep arousal and sleep problems of anxiety-disordered youth. Child Psychiatry & Human Development, 41, 156–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alkon A, Boyce WT, Neilands TB, & Eskenazi B (2017). Children’s Autonomic Nervous System Reactivity Moderates the Relations between Family Adversity and Sleep Problems in Latino 5-Year Olds in the CHAMACOS Study. Frontiers in Public Health, 5, 1–11. doi: 10.3389/fpubh.2017.00155 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bagley EJ, Kelly RJ, Buckhalt JA, & Sheikh El, M. (2015). What keeps low-SES children from sleeping well: The role of presleep worries and sleep environment. Sleep Medicine, 16, 496–502. 10.1016/j.sleep.2014.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bonnet MH, & Arand DL (2010). Hyperarousal and insomnia: State of the science. Sleep Medicine Reviews, 14, 9–15. doi: 10.1016/j.smrv.2009.05.002 [DOI] [PubMed] [Google Scholar]
  5. Critchley HD (2002). Electrodermal Responses: What Happens in the Brain. The Neuroscientist, 8, 132–142. [DOI] [PubMed] [Google Scholar]
  6. Crowley SJ (2013). Assessment of Circadian Rhythms. In Wolfson AR & Montgomery-Downs HE (Eds.), Oxford Handbook of Infant, Child, and Adolescent Sleep and Behavior (pp. 204–222). [Google Scholar]
  7. Dahl RE (1996). The regulation of sleep and arousal: Development and psychopathology. Development and Psychopathology. Special Issue: Reflecting on the Past and Planning for the Future of Developmental Psychopathology, 8, 3–27. doi: 10.1017/s0954579400006945 [DOI] [Google Scholar]
  8. Dawson ME, Schell AM, & Filion DL (2000). The Electrodermal System. In Cacioppo JT, Tassinary LG, & Berntson GG (Eds.), Handbook of Psychophysiology (2nd ed., pp. 200–223). [Google Scholar]
  9. Delannoy J, Mandai O, Honoré J, Kobayashi T, & Sequeira H (2015). Diurnal Emotional States Impact the Sleep Course. PLoS One, 10, 1–18. doi: 10.1371/journal.pone.0142721 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. El-Sheikh M, & Buckhalt JA (2005). Vagal regulation and emotional intensity predict children’s sleep problems. Developmental Psychobiology, 46, 307–317. doi: 10.1002/dev.20066 [DOI] [PubMed] [Google Scholar]
  11. El-Sheikh M, & Arsiwalla DD (2010). Children’s sleep, skin conductance level and mental health. Journal of Sleep Research, 20, 326–337. doi: 10.1111/j.1365-2869.2010.00880.x [DOI] [PubMed] [Google Scholar]
  12. El-Sheikh M, Erath SA, & Bagley EJ (2013). Parasympathetic nervous system activity and children’s sleep. Journal of Sleep Research, 22(3), 282–288. doi: 10.1111/jsr.12019 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Elmore-Staton L, El-Sheikh M, Vaughn B, & Arsiwalla DD (2012). Preschoolers’ daytime respiratory sinus arrhythmia and nighttime sleep. Physiology and Behavior, 107(3), 414–417. doi: 10.1016/j.physbeh.2012.07.005 [DOI] [PubMed] [Google Scholar]
  14. Espie CA (2002). Conceptual Issues in the Development, Persistence, and Treatment of Sleep Disorder in Adults. Annual Review of Psychology, 53, 215–243. [DOI] [PubMed] [Google Scholar]
  15. Farina B, Dittoni S, Colicchio S, Testani E, Losurdo A, Gnoni V, et al. (2013). Heart Rate and Heart Rate Variability Modification in Chronic Insomnia Patients. Behavioral Sleep Medicine, 12, 290–306. doi: 10.1080/15402002.2013.801346 [DOI] [PubMed] [Google Scholar]
  16. Fernandez-Mendoza J (2017). The insomnia with short sleep duration phenotype. Current Opinion in Psychiatry, 30, 56–63. doi: 10.1097/YCO.0000000000000292 [DOI] [PubMed] [Google Scholar]
  17. Fisher BE, Ross K, & Wilson A (1994). Parents“ and Children”s Ratings of Sleep Behavior, Excitement, and Tiredness: A 10-Week Longitudinal Study. The Journal of Genetic Psychology, 155, 191–199. doi: 10.1080/00221325.1994.9914771 [DOI] [PubMed] [Google Scholar]
  18. Galland BC, Taylor BJ, Elder DE, & Herbison P (2012). Normal sleep patterns in infants and children: a systematic review of observational studies. Sleep Medicine Reviews, 16, 213–222. doi: 10.1016/j.smrv.2011.06.001 [DOI] [PubMed] [Google Scholar]
  19. Garde AH, Albertsen K, Persson R, Hansen ÅM, & Rugulies R (2012). Bi-Directional Associations Between Psychological Arousal, Cortisol, and Sleep. Behavioral Sleep Medicine, 10, 28–40. doi: 10.1080/15402002.2012.636272 [DOI] [PubMed] [Google Scholar]
  20. Gilbert SS, van den Heuvel CJ, Ferguson SA, & Dawson D (2004). Thermoregulation as a sleep signaling system. Sleep Medicine Reviews, 8, 81–93. doi: 10.1016/S1087-0792(03)00023-6 [DOI] [PubMed] [Google Scholar]
  21. Gregory AM, & O’Connor TG (2002). Sleep problems in childhood: A longitudinal study of developmental change and association with behavioral problems. Journal of the American Academy of Child & Adolescent Psychiatry, 41, 964–971. [DOI] [PubMed] [Google Scholar]
  22. Gregory AM, Willis TA, Wiggs L, Harvey AG, & the STEPS Team (2008). Presleep Arousal and Sleep Disturbances in Children. Sleep, 31, 1745–1747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Harrell FE Jr (2019). rms: Regression Modeling Strategies. R package version 5.1–4. https://CRAN.R-project.org/package=rms [Google Scholar]
  24. Herlan A, Ottenbacher J, Schneider J, Riemann D, & Feige B (2018). Electrodermal activity patterns in sleep stages and their utility for sleep versus wake classification. Journal of Sleep Research, 28, 1–8. doi: 10.1111/jsr.12694 [DOI] [PubMed] [Google Scholar]
  25. Hollingshead AB (1975). Four factor index of social status. Unpublished work. [Google Scholar]
  26. Hot P, Naveteur J, Leconte P, & Sequeira H (1999). Diurnal variations of tonic electrodermal activity. International Journal of Psychophysiology, 33, 223–230. [DOI] [PubMed] [Google Scholar]
  27. Huber PJ (1967). The behavior of maximum likelihood estimates under nonstandard conditions. Paper presented at the Proceedings of the fifth Berkeley symposium on mathematical statistics and probability. [Google Scholar]
  28. Kleckner IR, Jones RM, Wilder-Smith O, Wormwood JB, Akcakaya M, Quigley KS, et al. (2018). Simple, Transparent, and Flexible Automated Quality Assessment Procedures for Ambulatory Electrodermal Activity Data. IEEE Transactions on Biomedical Engineering, 65, 1460–1467. doi: 10.1109/TBME.2017.2758643 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lack LC, Gradisar M, Van Someren EJW, Wright HR, & Lushington K (2008). The relationship between insomnia and body temperatures. Sleep Medicine Reviews, 12, 307–317. doi: 10.1016/j.smrv.2008.02.003 [DOI] [PubMed] [Google Scholar]
  30. Lester BK, Burch NR, & Dossett RC (1967). Nocturnal EEG-GSR Profiles: The Influence of Presleep States. Psychophysiology, 3, 238–248. [DOI] [PubMed] [Google Scholar]
  31. Meltzer LJ, Montgomery-Downs HE, Insana SP, & Walsh CM (2012). Use of actigraphy for assessment in pediatric sleep research. Sleep Medicine Reviews, 16, 463–475. doi: 10.1016/j.smrv.2011.10.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mindell JA, Telofski LS, Wiegand B, & Kurtz ES (2009). A nightly bedtime routine: impact on sleep in young children and maternal mood. Sleep, 32, 599–606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mindell JA, & Williamson AA (2018). Benefits of a bedtime routine in young children: Sleep, development, and beyond. Sleep Medicine Reviews, 40, 93–108. doi: 10.1016/j.smrv.2017.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Monroe LJ (1967). Psychological and Physiological Differences between Good and Poor Sleepers. Journal of Abnormal Psychology, 72, 255–264. [DOI] [PubMed] [Google Scholar]
  35. Morin CM, Rodrigue S, & Ivers H (2003). Role of stress, arousal, and coping skills in primary insomnia. Psychosomatic Medicine, 65, 259–267. doi: 10.1097/01.psy.0000030391.09558.a3 [DOI] [PubMed] [Google Scholar]
  36. Nelson J, & Harvey AG (2003). An exploration of pre-sleep cognitive activity in insomnia: Imagery and verbal thought. British Journal of Clinical Psychology, 42, 271–288. doi: 10.1016/j.smrv.2011.02.0 05 [DOI] [PubMed] [Google Scholar]
  37. Nicassio PM, Mendlowitz DR, Fussell JJ, & Petras L (1985). The phenomenology of the pre-sleep state: The development of the pre-sleep arousal scale. Behaviour Research and Therapy, 23, 263–271. [DOI] [PubMed] [Google Scholar]
  38. Nikula R (1991). Psychological Correlates of Nonspecific Skin Conductance Responses. Psychophysiology, 28, 86–90. [DOI] [PubMed] [Google Scholar]
  39. Okamoto-Mizuno K, Mizuno K, & Shirakawa S (2016). Sleep and Skin Temperature in Preschool Children and Their Mothers. Behavioral Sleep Medicine, 16, 64–78. doi: 10.1080/15402002.2016.1173552 [DOI] [PubMed] [Google Scholar]
  40. Patriquin MA, Mellman TA, Glaze DG, & Alfano CA (2014). Polysomnographic sleep characteristics of generally-anxious and healthy children assessed in the home environment. Journal of Affective Disorders, 161, 79–83. doi: 10.1016/j.jad.2014.02.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. R Development Core Team (2014). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/. [Google Scholar]
  42. Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, & Nissen C (2010). The hyperarousal model of insomnia: A review of the concept and its evidence. Sleep Medicine Reviews, 14, 19–31. doi: 10.1016/j.smrv.2009.04.002 [DOI] [PubMed] [Google Scholar]
  43. Sadeh A, Sharkey KM, & Carskadon MA (1994). Activity-based sleep—wake identification: An empirical test of methodological issues. Sleep, 17, 201–207. [DOI] [PubMed] [Google Scholar]
  44. Schwichtenberg AJ, Abel EA, Keys E, & Honaker SM (2019). Diversity in pediatric behavioral sleep intervention studies. Sleep Medicine Reviews, 47, 103–111. doi: 10.1016/j.smrv.2019.07.004 [DOI] [PubMed] [Google Scholar]
  45. Schwichtenberg AJ, & O’Haire M (2017). Measuring electrodermal activity in children with autism: A data processing platform. Gadomski Foundation Grant, Purdue University, IN. [Google Scholar]
  46. Staples AD, Bates JE, Petersen IT, McQuillan ME, & Hoyniak C (2019). Measuring sleep in young children and their mothers: Identifying actigraphic sleep composites. International Journal of Behavioral Development, 43, 278–285. doi: 10.1177/0165025419830236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Stein PK, & Pu Y (2012). Heart rate variability, sleep and sleep disorders. Sleep Medicine Reviews, 16, 47–66. doi: 10.1016/j.smrv.2011.02.005 [DOI] [PubMed] [Google Scholar]
  48. Wang M (2014). Generalized Estimating Equations in Longitudinal Data Analysis: A Review and Recent Developments. Advances in Statistics, 2014, 1–11. doi: 10.1155/2014/303728 [DOI] [Google Scholar]
  49. Whalen DJ, Sylvester CM, & Luby JL (2017). Depression and anxiety in preschoolers. Child and Adolescent Psychiatric Clinics of North America, 26(3), 503–522. doi: 10.1016/j.chc.2017.02.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. White H (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica: Journal of the Econometric Society, 817–838. [Google Scholar]
  51. Wicklow A, & Espie CA (2000). Intrusive thoughts and their relationship to actigraphic measurement of sleep: Towards a cognitive model of insomnia. Behaviour Research and Therapy, 38, 679–693. [DOI] [PubMed] [Google Scholar]
  52. Williams KE, Berthelsen D, Walker S, & Nicholson JM (2017). A developmental cascade model of behavioral sleep problems and emotional and attentional self-regulation across early childhood. Behavioral Sleep Medicine, 15, 1–21. doi: 10.1080/15402002.2015.1065410 [DOI] [PubMed] [Google Scholar]
  53. Williamson AA, & Mindell JA (2020). Cumulative socio-demographic risk factors and sleep outcomes in early childhood. Sleep, 389, 77–13. doi: 10.1093/sleep/zsz233 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Winzeler K, Voellmin A, Schäfer V, Meyer AH, Cajochen C, Wilhelm FH, & Bader K (2014). Daily stress, presleep arousal, and sleep in healthy young women: a daily life computerized sleep diary and actigraphy study. Sleep Medicine, 15, 359–366. doi: 10.1016/j.sleep.2013.09.027 [DOI] [PubMed] [Google Scholar]
  55. Wuyts J, De Valck E, Vandekerckhove M, Pattyn N, Bulckaert A, Berckmans D, et al. (2012). The influence of pre-sleep cognitive arousal on sleep onset processes. International Journal of Psychophysiology, 83, 8–15. doi: 10.1016/j.ijpsycho.2011.09.016 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp 1
NIHMS1760198-supplement-Supp_1.docx (105.4KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, C.H., upon reasonable request.

RESOURCES