Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Nov 6.
Published in final edited form as: Dev Psychobiol. 2016 Aug 30;59(1):60–69. doi: 10.1002/dev.21467

Sleep duration and RSA suppression as predictors of internalizing and externalizing behaviors

Sunghye Cho 1,*, Lauren E Philbrook 2, Elizabeth L Davis 3, Kristin A Buss 1
PMCID: PMC5673473  NIHMSID: NIHMS916703  PMID: 27577700

Abstract

Although the conceptual interplay among the biological and clinical features of sleep, arousal, and emotion regulation has been noted, little is understood about how indices of sleep duration and parasympathetic reactivity operate jointly to predict adjustment in early childhood. Using a sample of 123 toddlers, the present study examined sleep duration and RSA reactivity as predictors of internalizing and externalizing behaviors. Parents reported on children’s sleep duration and adjustment. RSA reactivity was assessed via children’s responses to fear-eliciting stimuli and an inhibitory control challenge. Findings demonstrated that greater RSA suppression to both types of tasks in combination with longer sleep duration was concurrently associated with less internalizing. In contrast, greater RSA augmentation to an inhibitory control task in the context of shorter sleep duration predicted more externalizing 1 year later. The significance of duration of toddlers’ sleep as well as the context in which physiological regulatory difficulties occurs is discussed.

Keywords: internalizing and externalizing behaviors, respiratory sinus arrhythmia, sleep

1 INTRODUCTION

Sleep and physiological regulation are implicated in the regulation of affect and behavior (Dahl, 1996), and measures of both of these bioregulatory systems have been linked to concurrent or later risk for childhood psychopathology (Beauchaine, 2001; El-Sheikh, Kelly, Buckhalt, & Hinnant, 2010; Sadeh, 2007). During the first 3 years of life, children’s sleep duration declines significantly and self-regulatory competence improves at a rapid pace. The reorganization of sleep and self-regulation occurring during this developmental period suggests that it is a critical time for studying these factors that may influence trajectories of adjustment across early childhood. The present study examined the joint and independent contributions of sleep and physiological regulation to internalizing and externalizing behaviors by focusing on toddlers’ sleep duration and parasympathetic regulation.

1.1 Associations between child sleep duration and behavior

In general, sleep duration is relatively stable across the toddler period, though there are individual differences (Jenni, Molinari, Caflisch, & Largo, 2007). A dramatic developmental shift in sleep duration has been reported in early childhood, with a particularly marked decrease in nocturnal sleep duration documented between 1.5 and 2 years of age (Acebo et al., 2005). Following this decline in sleep duration, 2-year olds are estimated to obtain approximately 10.4 (Acebo et al., 2005) to 11.5 (Iglowstein, Jenni, Molinari, & Largo, 2003) hr of nightly sleep.

In comparison to research examining the behavioral correlates of childhood sleep problems (e.g., night wakings), the association between childhood sleep duration and behavioral functioning has received limited empirical attention (Sadeh, 2007; Staples & Bates, 2011). However, as outlined by Sadeh (2007), sufficient duration of sleep is thought to facilitate an alert daytime state that allows for active engagement with the environment. Thus, children who experience reduced sleep may have more difficulty modulating their arousal appropriately to context, which manifests as behavior problems. Supporting this contention, an experimental study found that 2–3-year olds deprived of a daytime nap exhibited more negativity in response to frustration and to negative pictures, and less positivity to solving a puzzle and to positive pictures, compared to when they were well-rested (Berger, Miller, Seifer, Cares, & Lebourgeois, 2012).

Although the amount of sleep associated with optimal daytime functioning in childhood remains not well understood, it has been suggested that 10 hr represents a lower threshold value for children between 2 and 3 years of age. For instance, Lavigne et al. (1999) reported that 2–3-year olds sleeping <10 hr at night exhibited elevated levels of externalizing behaviors. Similarly, Touchette et al. (2007) found that maternal report of reduced sleep duration (i.e., < 10 hr per night) at 2.5 years of age predicted externalizing behaviors such as hyperactivity and impulsivity at 6 years, even when sleep duration increased later in development for these children. Fewer studies have examined sleep duration and internalizing problems among children in this age group. Although one study has indicated that 2-year-old toddlers sleeping <12.5 hr were more likely to develop anxiety and depressive symptoms at age 3 (Jansen et al., 2011), another study did not find a significant association between sleep amount and internalizing problems for 2–3-year olds (Lavigne et al., 1999). The potentially adverse effects of sleeping significantly longer than age norms on daytime behavioral functioning in early childhood remains largely unexplored, with one study indicating that parent report of child’s longer sleep duration is associated with behavioral aggression (Coulombe, Reid, Boyle, & Racine, 2010).

1.2 Associations between child parasympathetic regulation and behavior

According to Porges’ Polyvagal theory, the mammalian parasympathetic nervous system (PNS), coordinated by the myelinated (“smart”) vagus nerve, is responsible for modulating cardiac output in support of regulation of emotion, attention, and behavior (Porges, 2007). Respiratory sinus arrhythmia (RSA), a measure of the variability in heart rate within the range of spontaneous respiration, is commonly used as an index of parasympathetic regulation (Berntson et al., 1997). Higher baseline RSA, indicative of greater vagal influence on the heart, may reflect a greater ability to engage with the environment and to cope with environmental challenge (Porges, 2007). Suppression of parasympathetic influence over the heart under conditions of challenge (a decrease in RSA relative to basal levels) is theorized to allow for increases in cardiac output that facilitate the individual’s coping responses to environmental challenges. In contrast, an increase, or augmentation of parasympathetic function, returns the system to homeostasis (Porges, 2007). Theory suggests that appropriate physiological regulation of arousal supports children’s ability to effectively manage their emotions, which in turn facilitates social competence (Beauchaine, 2001). RSA suppression in response to challenge has been linked to better emotion regulation and lower levels of behavior problems, negativity, aggression, and social withdrawal as well as better social skills in young children (Calkins & Dedmon, 2000; Calkins & Keane, 2004; Porges, Doussard-Roosevelt, Portales, & Greenspan, 1996). RSA augmentation to challenge, in contrast, has been associated with internalizing problems (Heilman et al., 2012) including fearfulness (Liew et al., 2011) as well as externalizing behavior problems (Calkins, Graziano, & Keane, 2007). In the present study, children’s RSA responses to fear and delay of gratification tasks were examined because of their correspondence with the behavior problems of interest; children at risk for internalizing may have more difficulty regulating in response to fear stimuli (Brooker et al., 2013), whereas for children at risk for externalizing a task requiring inhibitory control may be particularly challenging for physiological regulation (Sulik, Eisenberg, Spinrad, & Silva, 2015).

1.3 Evidence for interactions between child sleep and parasympathetic regulation

There has been growing interest in examining sleep and parasympathetic regulation as interrelated bioregulatory processes that jointly contribute to childhood behavioral problems. For example, a study by El-Sheikh and colleagues provided evidence that children’s RSA moderates the influence of sleep on behavioral outcomes (El-Sheikh, Erath, & Keller, 2007). Poorer parasympathetic regulation, as indicated by attenuated RSA suppression to challenge, may represent vulnerability for adjustment issues related to difficulty sleeping (El-Sheikh et al., 2007). This is also consistent with the diathesis–stress model (Monroe & Simons, 1991; Zuckerman, 1979), in which experiences of stress are thought to activate vulnerability factors (i.e., diatheses) and place children at risk. It has been theorized that children who have more difficulty physiologically regulating and who sleep less may be more likely to develop reactive stress responses (Dahl, 1996), which may in turn increase risk for both internalizing and externalizing types of behaviors (El-Sheikh et al., 2007; El-Sheikh & Arsiwalla, 2011)

To our knowledge, only one previous study examined this interactive influence in children, and found that 8–9-year olds who slept less during the night reported more depressive symptoms and externalizing behaviors when they also had lower basal RSA (El-Sheikh et al., 2007). Children who slept less also were at increased risk for externalizing when they showed less RSA suppression to emotional challenge. Similarly, El-Sheikh and Arsiwalla (2011) found that reduced sleep duration predicted more depressive symptoms when children had lower basal skin conductance and less skin conductance reactivity to an attentional challenge. Together, these studies suggest that in the context of less sleep, poorer physiological regulation may increase risk for behavioral symptoms.

2 CURRENT STUDY

The goal of the current study was to characterize the unique and interactive influences of sleep duration and parasympathetic regulation (RSA reactivity to challenges) on behavioral outcomes in a toddler sample. Data for the current study were drawn from a larger longitudinal study of temperament and socio-emotional development, for which inhibited children were over-selected. Our sample included all children who participated in the larger study at ages 2 and 3. We aimed to build upon the existing literature in several ways. First, because inhibited toddlers were over-represented in the sample, it was uniquely suited for studying predictors of behavior problems, particularly internalizing. Second, sleep duration was examined as a predictor of behavior, an association that is less commonly studied. Based on extant research indicating that reduced sleep duration is related to greater behavioral problems, it was hypothesized that less sleep would be associated with both concurrent and longitudinal behavioral problems. To further advance the literature on sleep duration, we sought to identify distinct ranges of sleep that may be associated with both internalizing and externalizing behavior. Third, the interactive influence of sleep and RSA on behavior, which to our knowledge has not been previously tested in toddlers, was examined concurrently and longitudinally.

In examining children’s RSA, the role of context was also explored, wherein changes in RSA in response to different types of stimuli (fear, inhibitory control) were compared as predictors of behavior. It was hypothesized that reduced sleep duration and less RSA suppression would jointly predict behavioral problems. Furthermore, context-specific effects were anticipated, such that less RSA suppression to fear-eliciting contexts would be closely associated with greater internalizing symptoms and less RSA suppression to the inhibitory control task would be closely associated with greater externalizing symptoms.

3 METHODS

3.1 Participants

The present study utilized data from 123 children (60 girls) who participated in a longitudinal study of temperament and socio-emotional development. Families with toddlers between the ages of 18 and 20 months were recruited using public birth announcements published in a semi-rural northeastern city and its surrounding communities. Eligible families that returned study consent forms were asked to complete screening questionnaires. Due to the larger study’s focus on the development of inhibition in childhood, these screening measures were used to identify and enroll an over-sample of fearful toddlers (n = 63) scoring at least 1 standard deviation above the mean on parent ratings of fearfulness and wariness. The remaining children in the sample scored below this cutoff. The majority of the sample was middle class (M = 49.72, SD = 10.70 on the Hollingshead index) with family income (assessed at age 2) ranging from $15,000 or less to more than $60,000. The sample was primarily Caucasian (n = 113; 90.4%), followed by Asian-American (n = 8; 6.4%), American-Indian (n = 2; 1.6%), African-American (n = 1; .8%), and Hispanic (n = 1; .8%).

3.2 Procedure

At age 2 (Mage = 24.43 months, SD =.47), toddlers’ cardiovascular data were obtained during a series of laboratory episodes designed to assess children’s fearfulness and inhibitory control. After the lab assessment, parents completed a sleep diary for a 1-week period regarding the onset and duration of children’s sleep. Parents also completed questionnaires regarding children’s internalizing and externalizing as part of the age 2 assessment. Parents were mailed these same measures to complete when children were 3.

3.2.1 Age 2 laboratory visit

At the age 2 laboratory visit, toddlers participated in a series of tasks including a 5-min baseline, modified versions of the Laboratory Temperament Assessment Battery (Buss & Goldsmith, 2000), and the Snack Delay task (Kochanska, Murray, Jacques, Koenig, & Vandegeest, 1996). Prior to the baseline, cardiac electrodes were placed on the toddlers’ torso and back. Leads connected these electrodes to a personal digital assistant (PDA) device that wirelessly transmitted the information to a computer in the next room. The child wore the PDA in a small backpack throughout data collection. After the experimenters visually inspected the collection of the ECG data, toddlers engaged in quiet activities with an experimenter (e.g., coloring, reading a book) for the 5-minute baseline period. The present study examined cardiovascular data collected during six laboratory episodes designed to examine individual differences in children’s fear responses (Lab-TAB; Buss & Goldsmith, 2000). These episodes—hereafter referred to as “threat episodes”—included an interactive puppet show (Puppet Show, 3 min), playing with a female experimenter dressed as a clown (Clown, 3 min), social interaction with strangers (Stranger Approach, 1.5 min, and Stranger Working, 2 min), and two object fear tasks that involved presentation of remote controlled toy spider and robot (Spider, 1 min and Robot, 1 min; Buss, 2011). As mentioned, cardiac data were also recorded during a modified version of the Snack Delay task (Kochanska et al., 1996). In this episode, the experimenter introduced a game with M & M candies. Children were seated at a child sized table and were instructed to wait to take the candies placed under a transparent cup until the experimenter signaled the approval by ringing a bell. A total of six trials were presented with varying time intervals (5 s, 10 s, no pause, 20 s, no pause, 30 s) and the entire task lasted 5 min. For all tasks, parents were seated in the room and asked to remain as uninvolved as possible, except if their child became highly distressed in which case they could comfort the child as they wished. The episodes were presented in the same order to each child, including four other episodes that were not included in the analyses for the current study because they measured other behavioral constructs. The task order was as follows: Puppet Show, Stranger Approach, Snack Delay, Robot, Clown, Stranger Working, & Spider.

3.3 Measures

3.3.1 Respiratory sinus arrhythmia (RSA)

Cardiac output measures were collected using the Mindware Wi-Fi ACQ software, Version 1.0. (Mindware Technologies, LTD, Westerville, OH) and were analyzed offline using the Mindware Heart Rate Variability version 2.51. Interbeat intervals (IBI) and improbable intervals were identified given the overall distribution. Data were detrended using a first-order polynomial. Fast Fourier Transformation of the edited data was performed to obtain the RSA scores.

RSA was defined as the natural log integral of the very high frequency .24- to 1.04-Hz power band (Fracasso, Porges, Lamb, & Rosenberg, 1994) and calculated in 30-s epochs. Visual inspection and editing of artifacts in the ECG data were completed by three scorers who achieved good interrater reliability (agreement = 86%) on 25% of the files. RSA during the baseline, six threat episodes, and Snack Delay were computed by averaging across the 30-s epochs. Average RSA values across the six threat episodes were highly correlated (rs = .55 to .81). Thus, an average RSA score across the six threat episodes was computed. We were interested in the change in children’s RSA from resting baseline to the threat episodes (averaged) and to Snack Delay. Two reactivity scores were calculated by regressing task RSA on baseline RSA and extracting the standardized residuals for use in analysis (i.e., residualized change scores that accounted for children’s initial RSA levels). Lower reactivity scores indicated greater RSA suppression from baseline to task relative to the sample average.

3.3.2 Sleep duration

As part of the age 2 assessment, caregivers completed a sleep diary and recorded bedtime, waketime, and duration of toddlers’ sleep for a 1-week period. Based on this report, the average nighttime sleep duration was computed for 5 weekdays. According to the extant literature, parental report of children’s sleep duration has shown adequate validity and reliability (Sadeh, 2004). Additionally, it has been suggested that evidence of long-term stability of sleep duration during childhood indicates that sleep duration may be considered as trait-like (Jenni et al., 2007).

3.3.3 Parent rating of behavior and regulatory problems (ages 2 and 3)

Caregivers completed the Infant-Toddler Social and Emotional Assessment—Revised (ITSEA; Carter, Briggs-Gowan, Jones, & Little, 2003) when children were ages 2 and 3. The ITSEA is a parent-report questionnaire designed to assess a wide range of socio-emotional problems and competencies in 1- to 3-year-old children. The Internalizing domain comprises subscales including Depression/Withdrawal, General Anxiety, Separation Distress, and Inhibition to Novelty. The Externalizing domain is divided into three subscales which include Activity/Impulsivity, Aggression/Defiance, and Peer Aggression scales. Additionally, the Sleep subscale of the Dysregulated Domain assesses sleep problems. Given that the present study focused on the specific contribution of sleep duration, the Sleep subscale was used only to statistically control for children’s co-occurring sleep problems at age 2. Acceptable reliability and validity have been demonstrated for the ITSEA (Briggs-Gowan & Carter, 1998, 2007). Good internal consistency was found for all of the scales that were examined in the present study: Internalizing (α = .82 to .84), Externalizing (α = .83 to .88), and Sleep (α = .89).

3.4 Analytical approach

3.4.1 Missing data

Examination of the missing data indicated that cardiac data (threat-episodes n = 29; Snack Delay n = 35) and age 3 questionnaire data (n = 23) were unavailable for <30% of the sample. Missing data were imputed using multiple imputation (MI), as MI has been shown to perform very well for data that is up to 50% missing (Graham & Schafer, 1999). Children with missing data did not differ on any of the demographic variables from those without missing data. Those with missing cardiac data were rated as lower in age 2 externalizing behaviors (F = 4.57, p = .04). As recommended by Graham (2009, 2012), age 2 externalizing was included as auxiliary variable in the imputation model to reduce the potential bias that would be introduced by attrition. The multivariate imputation by chained equations (MICE) package in R statistical software was employed to impute and to pool the results of the analyses (van Buuren & Groothuis-Oudshoorn, 2011).

3.4.2 Statistical analyses

Analyses proceeded in the following order: First, data were examined for normality and potential outliers. Two outlier data points that exceeded three standard deviations above the mean value of internalizing symptom scores were identified for two participants and were replaced with the next highest value in the data (Tabachnick & Fidell, 2007). Next, descriptive statistics and correlations were computed for all study variables. Finally, the concurrent and predictive relations among RSA, sleep duration, and symptoms of internalizing and externalizing behaviors at ages 2 and 3 were examined via a series of hierarchical regression analyses. All of the predictor variables were mean-centered prior to creating the interaction terms (Aiken & West, 1991). Significant interaction effects were probed using the tests of simple slopes (at 1 SD above and below mean) and were further examined to identify the points at which the association between RSA suppression and behavioral outcomes became significant (Dearing & Hamilton, 2006). Finally, the relative proportion of cases that were differentially influenced by distinct ranges of sleep duration was estimated using the proportion affected (PA) index (Roisman et al., 2012). The PA index computes the proportion of cases that fall below and above the point at which two simple slopes intersect (“cross over value”). Parent rating of age 2 sleep problems was entered as a covariate in all of the models to account for the potential contribution of sleep problems to our pattern of findings.

4 RESULTS

4.1 Descriptive statistics

Means, standard deviations, and correlations among the study variables can be found in Table 1. Sleep duration at age 2 was inversely associated with concurrent sleep dysregulation (r = −.34, p < .001), and with age 3 externalizing behaviors (r = −.26, p = .01). RSA suppression to threat and Snack Delay were positively associated (r = .69, p < .001). RSA suppression to threat was also associated with age 3 internalizing behaviors (r = .24, p = .04), indicating that toddlers who demonstrated less RSA suppression to threat were more likely to develop internalizing behaviors a year later. Gender differences in key study variables were not found. As such, gender was not included in subsequent analyses.

TABLE 1.

Means, standard deviations, and bivariate correlations of study variables

Variable Mean SD 1 2 3 4 5 6 7 8
Age 2 variables
 1. Sleep duration 11.00 .96 –
 2. Baseline RSA 4.46 1.08 .05 –
 3. RSA TE 4.39 .98 −.04 .00 –
 4. RSA SD 4.36 1.06 −.00 .00 .69*** –
 5. Internalizing .42 .18 −.15 .08 .07 .12 –
 6. Externalizing .44 .23 −.14 −.05 .01 −.02 .09 –
 7. Sleep problems .47 .54 −.34*** .15 .05 .06 .40*** .26** –
Age 3 variables
 8. Internalizing .41 .22 −.14 .12 .24* .09 .69*** .06 .23* –
 9. Externalizing .48 .23 −.26** −.07 .05 −.09 .07 .66*** .22* .12
Open in a new tab

Note. RSA, respiratory sinus arrhythmia; RSA TE, residualized change score of RSA from baseline to six threat episodes; RSA SD, residualized change score of RSA from baseline to Snack Delay.

*

p < .05.

**

p < .01.

***

p < .001.

4.2 Hierarchical regression analyses

4.2.1 Prediction of age 2 internalizing behaviors

Sleep duration X RSA reactivity to threat

A significant interaction between sleep duration and RSA suppression was present in a model that includedage2 internalizing symptoms as the dependent variable (b = .23, p = .02) (Table 2). Probing this interaction effect revealed that longer sleep duration was associated with fewer internalizing symptoms for toddlers demonstrating a greater RSA suppression response (−1 SD) to threat (t = −2.71, p = .01). The association between sleep duration and internalizing symptoms was not significant in the context of less RSA suppression to threat tasks (+1 SD). Figure 1 illustrates the regions of significance, with the upper threshold of sleep duration occurring at approximately .24 SD below the mean (10.76 hr). The lower threshold of −8.13 SD (3.16 hr) exceeded the range of the present data. The two simple slopes crossed at 1.06 SD below the mean and approximately 86% of the sample fell above this upper threshold. Given that only the upper threshold of sleep duration was considered viable, the findings indicated that greater RSA suppression to threat was associated with fewer concurrent internalizing behaviors for a greater percentage of this sample.

TABLE 2.

Models predicting inhibition to novelty, internalizing, and externalizing symptoms

Internalizing (age 2)
Externalizing (age 2)
R2 F b R2 F b
Step 1 .10* 3.81* – .03 1.09 –
 Sleep – – −.02 – – –.06
 RSA TE – – .22* – – –.02
 Sleep X RSA TE – – .23* – – –.09
Step 2 .24** 8.59** .08† 2.28†
 Sleep problems – – .39** – – .23*
Step 1 .11* 4.10* – .03 1.19 –
 Sleep – – −.02 – – −.06
 RSA SD – – .19* – – −.07
 Sleep X RSA SD – – .24* – – −.08
Step 2 .23** 8.33** .08 1.68 –
 Sleep problems – – .37** – – .24*
Internalizing (age 3) Externalizing (age 3)
Step 1 .06 1.75 – .07† 2.59† –
 Sleep – – −.05 – – −.18†
 RSA TE – – .19† – – −.03
 Sleep X RSA TE – – .01 – – −.11
Step 2 .09† 2.42† – .09† 2.69†
 Sleep problems – – .18† .15
Step 1 .03 .82 – .11* 3.95* –
 Sleep – – −.06 – – −.18†
 RSA SD – – .07 – – .01
 Sleep X RSA SD – – –.05 – – −.23*
Step 2 .07† 2.60† – .13* 3.95* –
 Sleep problems – – .18† – – .16†
Open in a new tab

Note. RSA, respiratory sinus arrhythmia; RSA TE, residualized change score of RSA from baseline to six threat episodes; RSA SD, residualized change score of RSA from baseline to Snack Delay. Boldface indicates p-values < .05.

†

p < .10.

*

p < .05.

**

p < .01.

FIGURE 1.

FIGURE 1

Open in a new tab

RSA suppression as a moderator of the association between sleep duration and toddler internalizing and externalizing symptoms. The shaded areas represent regions of significance associated with age 2 sleep duration. (A) RSA suppression to threat as a moderator of the association between sleep duration and age 2 internalizing symptoms, (B) RSA suppression to Snack Delay as a moderator of the association between sleep duration and age 2 internalizing symptoms, (C) RSA suppression to Snack Delay episode as a moderator of the association between sleep duration and age 3 externalizing symptoms

Sleep duration X RSA reactivity to inhibitory control

A significant interaction effect of sleep duration and RSA suppression to the snack delay episode was related to concurrent internalizing behaviors (b = .24, p = .01). Testing the simple slopes (±1 SD) indicated that longer sleep duration was associated with fewer internalizing problems, only for toddlers showing greater RSA suppression (−1 SD) (t = −3.21, p = .002) (Figure 1). The simple slope at 1 SD above the mean for RSA suppression (i.e., less RSA suppression) was not significant. The interaction effect became significant below −2.61 SD (8.48 hr) and above −.15 SD (10.86 hr) relative to the mean value of sleep duration. Similar to the findings noted for RSA reactivity to threat, the lower threshold value exceeded the range of values for this data. Approximately 79% of toddlers were found to fall over the crossover point of 10.23 hr of sleep. Taken together, the inverse association between sleep duration and internalizing behaviors was significant only for the majority of toddlers who slept more than 10.86 hr per night and demonstrated greater RSA suppression to inhibitory control demand.

4.2.2 Prediction of age 3 internalizing behaviors

There was no evidence of sleep duration predicting age 3 internalizing behaviors independently or interactively with RSA suppression to threat or Snack Delay.

4.2.3 Prediction of age 2 externalizing behaviors

No significant main or interaction effects of sleep duration and RSA suppression to threat were found. Likewise, the model including RSA suppression to Snack Delay as a predictor of age 2 externalizing behavior did not reveal a significant main effect nor an interaction effect with sleep duration, beyond what was found for concurrent sleep problems (b = .24, p = .01).

4.2.4 Prediction of age 3 externalizing behaviors

Sleep duration X RSA reactivity to threat

There was no evidence that sleep duration or RSA suppression to threat predicted age 3 externalizing behaviors.

Sleep duration X RSA reactivity to inhibitory control

A main effect of age 2 sleep duration was subsumed by a significant interaction between sleep duration and RSA suppression to Snack Delay predicting externalizing behaviors a year later (b = −.23, p = .03). Probing this interaction effect revealed that shorter sleep duration was associated with greater age 3 externalizing behaviors for toddlers who exhibited less RSA suppression (+1 SD) during the Snack Delay task (t = −3.64, p < .001). The simple slope tested at 1 SD below the mean value of RSA suppression was not significant. Test of regions of significance identified a lower threshold of .88 SD below the mean and an upper threshold of 1.83 SD above the mean for sleep duration. Therefore, less RSA suppression during the Snack Delay task (+1 SD) predicted greater externalizing behaviors for toddlers who slept <10.15 hr. In contrast, sleeping more than 12.77 hr was associated with fewer externalizing behaviors for toddlers showing a less pronounced RSA suppression response. As such, sleep duration predicted externalizing behaviors for toddlers, but only in the context of less RSA suppression to the task of inhibitory control. In this model, the simple slopes were found to cross near the mean (+.06 SD relative to the mean) at 11.05 hr, with approximately 52% of the toddlers falling below this value. As can be seen in Figure 1, the significant inverse association between sleep duration and externalizing behaviors thus influenced a greater proportion of toddlers who obtained <10.15 hr of nighttime sleep.

5 DISCUSSION

The present study examined sleep duration and RSA reactivity as indices of toddlers’ capacity for regulation and their relations to internalizing and externalizing behaviors. Consistent with prior research, obtaining approximately 10-hr of nighttime sleep at age 2 was generally associated with better behavioral outcomes (e.g., Lavigne et al., 1999; Touchette et al., 2007). Extending the literature on sleep and adjustment, the present study provides new evidence that toddlers’ nighttime sleep duration and RSA reactivity are jointly related to behavioral symptoms that develop in early childhood. Notably, the patterns of interaction were found to differ for internalizing and externalizing, highlighting the importance of considering the contexts in which toddlers’ capacity for regulation is observed.

5.1 Sleep duration, parasympathetic regulation, and internalizing behaviors

We found that longer sleep duration was associated with fewer concurrent symptoms of internalizing behaviors, but this was qualified by parasympathetic regulation; only children who showed high levels of RSA suppression to both threat and inhibitory control demand tasks showed this relation. Contrary to our prediction, shorter sleep duration did not interact with parasympathetic regulation to predict toddlers’ concurrent or future internalizing symptoms. The lack of a link between reduced sleep duration and internalizing outcomes in the present study is consistent with previous research that did not document a direct association between sleep restriction and concurrent internalizing problems in 2- to 3-year-old toddlers (Lavigne et al., 1999). To our knowledge, only one study has demonstrated a predictive relation between shorter nighttime sleep duration and internalizing problems one year later in toddlers (Jansen et al., 2011). The above study also reported a significant main effect of the frequency of night wakings, yet it was unclear whether sleep duration may uniquely predict internalizing problems a year later. Although research has indicated that childhood sleep problems may precede the onset of internalizing disorders (Alvaro, Roberts, & Harris, 2013; Gregory et al., 2005; Gregory, Rijsdijk, Lau, Dahl, & Eley, 2009), more research is clearly needed to make definitive conclusions regarding the direction of influence linking sleep duration in early childhood and internalizing outcomes.

5.2 Sleep duration, parasympathetic regulation, and externalizing behaviors

We did not find associations between sleep or parasympathetic regulation and concurrent externalizing behaviors in toddlers, in contrast to previous work. For instance, Lavigne et al. (1999) reported that fewer than 10 hr of nighttime sleep was associated with concurrent externalizing behavior problems for 2–3-year-old toddlers. Although this finding underscores the importance of sleep duration in predicting externalizing outcomes, the effect of reduced sleep was not examined in conjunction with other forms of sleep problems (e.g., nightwaking) that are particularly common in toddlerhood. Given that we found no association between reduced sleep duration and concurrent externalizing behaviors beyond the contribution of concurrent sleep problems, examining the contributions of various sleep parameters in future developmental research will provide additional insight.

Consistent with a large body of literature linking reduced sleep to externalizing behaviors (Bates, Viken, Alexander, Beyers, & Stockton, 2002; Lavigne et al., 1999) such as hyperactivity (Touchette et al., 2007, 2009) and rule-breaking behaviors (Pesonen et al., 2010), reduced sleep at age two was associated with greater externalizing behavior problems a year later. Specifically, obtaining <10 hr of nighttime sleep predicted greater daytime externalizing behaviors a year later, but only in the context of poor physiological regulation during an inhibitory control task. Conversely, for a comparatively smaller proportion of toddlers, sleeping over 12.8 hr appeared to compensate for risk of later externalizing associated with poor physiological regulation. Using a nationally representative Canadian sample, Touchette and colleagues found that toddlers who were obtaining the least amount of sleep at 2.5 years of age were at greatest risk for developing hyperactive/impulsive behavior problems at six years of age. Therefore, our findings are consistent with other work showing that reduced sleep prior to three years of age uniquely predicts later hyperactive and impulsive behavioral problems.

5.3 Sleep and RSA reactivity: differential links to internalizing and externalizing behaviors

Insufficient sleep has been theorized to influence the regulation of affect (e.g., depressed mood, irritability, anger) or to increase susceptibility to behavioral impulsivity or disinhibition (Dahl, 1996). In the present study, examination of the interplay between sleep duration and physiological regulation during affective (i.e., threat-related) and behavioral self-regulation (i.e., inhibitory control) tasks revealed a differential pattern of results pertaining to internalizing and externalizing behaviors. Toddlers’ inadequate sleep appeared to confer a specific risk for preschool-age externalizing behaviors, but only in the context of poor physiological regulation during the inhibitory control challenge task. The specificity of the interaction effect involving RSA reactivity to the inhibitory control task is in line with previous research documenting the link between RSA reactivity to self-regulation tasks (e.g., delay of gratification) and externalizing behaviors (e.g., Sulik et al., 2015; Willoughby, Kupersmidt, Voegler-Lee, & Bryant, 2011). There was no evidence that inadequate sleep and poor regulation during the affective challenge was associated with a similar risk for externalizing behaviors. It is possible, however, that the absence of an interaction between reduced sleep and physiological regulation to affective challenge task reflects the nature of the emotion (threat/fear/wariness) that was elicited by the tasks used in the present study. For instance, prior research has demonstrated that children’s reactivity to an anger-eliciting task predicts externalizing behaviors (e.g., Gatzke-Kopp, Greenberg, & Bierman, 2015). As such, use of affective tasks that evoke anger rather than fear may further inform whether physiological regulation during certain affective challenges may also moderate the link between sleep duration and externalizing symptoms.

Although the interplay of sleep duration and physiological regulation predicted increased risk for externalizing outcomes, their relations to internalizing outcomes differed. Most notably, adequate sleep duration and adaptive regulation of RSA during affective challenge were jointly related to fewer symptoms of internalizing behaviors. This finding builds upon prior research showing emotion-specific associations between RSA reactivity to fear-relevant tasks and internalizing outcomes (e.g., Fortunato, Gatzke-Kopp, & Ram, 2013) and provides support to the notion that regulation of sleep, arousal, and affect overlap in physiological as well as clinical domains (Dahl, 1996).

5.4 Limitations

The present study highlighted the specific conditions under which sleep duration and parasympathetic regulation may relate to symptoms of internalizing and externalizing behavior. Nevertheless, a few potential limitations are noted. Although caregivers are reliable reporters of specific indices of children’s sleep—duration and onset times in particular (Sadeh, Raviv, & Gruber, 2000; Tikotzky & Sadeh, 2001), our data on toddlers’ sleep schedule indicated very high stability across a 5-week day period, which may potentially reflect parent difficulties with capturing slight variability in actual wake and sleep times. This idea is supported by previous research using parent-reported sleep duration that also indicated a high degree of stability of sleep duration over a week, with the difference in average sleep duration between weekdays and weekends estimated to be approximately 10 min (Touchette et al., 2007). The present study also examined context-specific patterns of physiological regulation during affective and behavioral challenge tasks. As RSA during affective challenge was averaged across six distinct threat episodes, regulation to threat may be considered a more reliable measure of RSA than regulation to Snack Delay. However, the Snack Delay task was substantially longer than the individual threat episodes and generated 10 distinct epochs across which RSA was averaged. Finally, this sample was recruited specifically to allow examination of the developmental trajectories associated with temperamental fearfulness, which afforded us a unique opportunity to explore the role of early emerging sleep and self-regulatory difficulties in a sample of children who may be at elevated risk for behavior problems. However, the limited socio-demographic diversity in the present sample may limit the generalizability of the results and future work should investigate similar questions with more diverse prospective longitudinal samples.

6 CONCLUSION

The present study aimed to advance the literature on developmental risks associated with regulatory processes by examining how context-specific patterns of parasympathetic regulation may moderate the associations among sleep and a range of internalizing and externalizing behavioral problems. Findings indicate that internalizing and externalizing behavioral problems are differentially influenced by the interaction of sleep duration and RSA reactivity to distinct contextual challenges. The present study also highlights the importance of considering the joint contribution of the specific range of sleep and contextual influences on psychophysiological vulnerability to a range of behavioral adjustment outcomes.

Acknowledgments

Funding Information

This work was supported by NIMH (R01MH75750) to Kristin A. Buss.

Footnotes

CONFLICTS OF INTEREST

The authors declare that they have no conflict of interest.

References

  1. Acebo C, Sadeh A, Seifer R, Tzischinsky O, Hafer A, Carskadon MA. Sleep/wake patterns derived from activity monitoring and maternal report for healthy 1-to 5-year-old children. Sleep. 2005;28:1568. doi: 10.1093/sleep/28.12.1568. [DOI] [PubMed] [Google Scholar]
  2. Aiken LS, West SG. Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage Publications; 1991. [Google Scholar]
  3. Alvaro PK, Roberts RM, Harris JK. A systematic review assessing bidirectionality between sleep disturbances, anxiety, and depression. Sleep. 2013;36:1059–1068. doi: 10.5665/sleep.2810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bates JE, Viken RJ, Alexander DB, Beyers J, Stockton L. Sleep and adjustment in preschool children: Sleep diary reports by mothers relate to behavior reports by teachers. Child Development. 2002;73:62–74. doi: 10.1111/1467-8624.00392. [DOI] [PubMed] [Google Scholar]
  5. Beauchaine T. Vagal tone, development, and Gray’s motivational theory: Toward an integrated model of autonomic nervous system in psychopathology. Development and Psychopathology. 2001;13:183–214. doi: 10.1017/S0954579401002012. [DOI] [PubMed] [Google Scholar]
  6. Berger RH, Miller AL, Seifer R, Cares SR, Lebourgeois MK. Acute sleep restriction effects on emotion responses in 30- to 36-month old children. Journal of Sleep Research. 2012;21:235–246. doi: 10.1111/j.1365-2869.2011.00962.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Berntson GG, Bigger JT, Eckberg DL, Grossman P, Kaufmann PG, Malik M, … Van der Molen MW. Heart rate variability: Origins, methods, and interpretive caveats. Psychophysiology. 1997;34:623–648. doi: 10.1111/j.1469-8986.1997.tb02140.x. [DOI] [PubMed] [Google Scholar]
  8. Briggs-Gowan MJ, Carter AS. Preliminary acceptability and psychometrics of the Infant-Toddler Social and Emotional Assessment (ITSEA): A new adult-report questionnaire. Infant Mental Health Journal. 1998;19:422–445. doi: 10.1002/(SICI)1097-0355(199824)19:4<422::AID-IMHJ5>3.0.CO;2-U. [DOI] [Google Scholar]
  9. Briggs-Gowan MJ, Carter AS. Applying the infant-toddler social & emotional assessment (ITSEA) and brief-ITSEA in early intervention. Infant Mental Health Journal. 2007;28:564–583. doi: 10.1002/imhj.20154. [DOI] [PubMed] [Google Scholar]
  10. Brooker RJ, Buss KA, Lemery-Chalfant K, Aksan N, Davidson RJ, Goldsmith HH. The development of stranger fear in infancy and toddlerhood: Normative development, individual differences, antecedents, and outcomes. Developmental Science. 2013;16:864–878. doi: 10.1111/desc.12058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Buss KA. Which fearful toddlers should we worry about? Context, fear regulation, and anxiety risk. Developmental Psychology. 2011;47:804. doi: 10.1037/a0023227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Buss KA, Goldsmith HH. Manual and normative data for the Laboratory Temperament Assessment Battery–Toddler Version. Madison: University of Wisconsin; 2000. [Google Scholar]
  13. Calkins SD, Dedmon SE. Physiological and behavioral regulation in two-year-old children with aggressive/destructive behavior problems. Journal of Abnormal Child Psychology. 2000;28:103–118. doi: 10.1023/A:1005112912906. [DOI] [PubMed] [Google Scholar]
  14. Calkins SD, Graziano PA, Keane SP. Cardiac vagal regulation differentiates among children at risk for behavior problems. Biological Psychology. 2007;74:144–153. doi: 10.1016/j.biopsycho.2006.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Calkins SD, Keane SP. Cardiac vagal regulation across the preschool period: Stability, continuity, and implications for childhood adjustment. Developmental Psychobiology. 2004;45:101–112. doi: 10.1002/dev.20020. [DOI] [PubMed] [Google Scholar]
  16. Carter AS, Briggs-Gowan MJ, Jones SM, Little TD. The infant-Toddler social and emotional assessment (ITSEA): Factor structure, reliability, and validity. Journal of Abnormal Child Psychology. 2003;31:495–514. doi: 10.1023/A:1025449031360. [DOI] [PubMed] [Google Scholar]
  17. Coulombe JA, Reid GJ, Boyle MH, Racine Y. Concurrent associations among sleep problems, indicators of inadequate sleep, psychopathology, and shared risk factors in a population-based sample of healthy Ontario children. Journal of Pediatric Psychology. 2010;35:790–799. doi: 10.1093/jpepsy/jsp097. [DOI] [PubMed] [Google Scholar]
  18. Dahl RE. The regulation of sleep and arousal: Development and psychopathology. Development and Psychopathology. 1996;8:3–27. doi: 10.1017/S0954579400006945. [DOI] [Google Scholar]
  19. Dearing E, Hamilton LC. Contemporary advances and classic advice for analyzing mediating and moderating variables. Monographs of the Society for Research in Child Development. 2006;71:88–104. doi: 10.1111/j.1540-5834.2006.00406.x. [DOI] [Google Scholar]
  20. El-Sheikh M, Arsiwalla DD. Children’s sleep, skin conductance level and mental health. Journal of Sleep Research. 2011;20:326–337. doi: 10.1111/j.1365-2869.2010.00880.x. [DOI] [PubMed] [Google Scholar]
  21. El-Sheikh M, Erath SA, Keller PS. Children’s sleep and adjustment: The moderating role of vagal regulation. Journal of Sleep Research. 2007;16:396–405. doi: 10.1111/j.1365-2869.2007.00618.x. [DOI] [PubMed] [Google Scholar]
  22. El-Sheikh M, Kelly RJ, Buckhalt JA, Hinnant JB. Children’s sleep and adjustment over time: The role of socioeconomic context. Child Development. 2010;81:870–883. doi: 10.1111/j.1467-8624.2010.01439.x. [DOI] [PubMed] [Google Scholar]
  23. Fortunato CK, Gatzke-Kopp LM, Ram N. Associations between respiratory sinus arrhythmia reactivity and internalizing and externalizing symptoms are emotion specific. Cognitive, Affective, & Behavioral Neuroscience. 2013;13:238–251. doi: 10.3758/s13415-012-0136-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fracasso M, Porges S, Lamb M, Rosenberg A. Cardiac activity in infancy: Reliability and stability of individual differences. Infant Behavior and Development. 1994;17:277–284. doi: 10.1016/0163-6383(94)90006-X. [DOI] [Google Scholar]
  25. Gatzke-Kopp LM, Greenberg M, Bierman K. Children’s parasympathetic reactivity to specific emotions moderates response to intervention for early-onset aggression. Journal of Clinical Child & Adolescent Psychology. 2015;44:291–304. doi: 10.1080/15374416.2013.862801. [DOI] [PubMed] [Google Scholar]
  26. Graham JW. Missing data analysis: Making it work in the real world. Annual Review of Psychology. 2009;60:549–576. doi: 10.1146/annurev.psych.58.110405.085530. [DOI] [PubMed] [Google Scholar]
  27. Graham JW. Missing data: Analysis and design. New York, NY: Springer; 2012. [DOI] [Google Scholar]
  28. Graham JW, Schafer JL. On the performance of multiple imputation for multivariate data with small sample size. In: Hoyle R, editor. Statistical strategies for small sample research. Thousand Oaks, CA: Sage; 1999. pp. 1–29. [Google Scholar]
  29. Gregory AM, Caspi A, Eley TC, Moffitt TE, O’Connor TG, Poulton R. Prospective longitudinal associations between persistent sleep problems in childhood and anxiety and depression disorders in adulthood. Journal of Abnormal Child Psychology. 2005;33:157–163. doi: 10.1007/s10802-005-1824-0. [DOI] [PubMed] [Google Scholar]
  30. Gregory AM, Rijsdijk FV, Lau JY, Dahl RE, Eley TC. The direction of longitudinal associations between sleep problems and depression symptoms: A study of twins aged 8 and 10 years. Sleep. 2009;32:189–199. doi: 10.1093/sleep/32.2.189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Heilman KJ, Connolly SD, Padilla WO, Wrzosek MI, Graczyk PA, Porges SW. Sluggish vagal brake reactivity to physical exercise challenge in children with selective mutism. Development and Psychopathology. 2012;24:241–250. doi: 10.1017/S0954579411000800. [DOI] [PubMed] [Google Scholar]
  32. Iglowstein I, Jenni OG, Molinari L, Largo RH. Sleep duration from infancy to adolescence: Reference values and generational trends. Pediatrics. 2003;111:302–307. doi: 10.1542/peds.111.2.302. [DOI] [PubMed] [Google Scholar]
  33. Jansen PW, Saridjan NS, Hofman A, Jaddoe VWV, Verhulst FC, Tiemeier H. Does disturbed sleeping precede symptoms of anxiety or depression in toddlers? The Generation R Study. Psychosomatic Medicine. 2011;73:242–249. doi: 10.1097/PSY.0b013e31820a4abb. [DOI] [PubMed] [Google Scholar]
  34. Jenni OG, Molinari L, Caflisch JA, Largo RH. Sleep duration from ages 1 to 10 years: Variability and stability in comparison with growth. Pediatrics. 2007;120:e769–e776. doi: 10.1542/peds.2006-3300. [DOI] [PubMed] [Google Scholar]
  35. Kochanska G, Murray K, Jacques TY, Koenig AL, Vandegeest KA. Inhibitory control in young children and its role in emerging internalization. Child Development. 1996;67:490–507. doi: 10.1111/j.1467-8624.1996.tb01747.x. [DOI] [PubMed] [Google Scholar]
  36. Lavigne JV, Arend R, Rosenbaum D, Smith A, Weissbluth M, Binns HJ, Christoffel KK. Sleep and behavior problems among preschoolers. Journal of Developmental & Behavioral Pediatrics. 1999;20:164–169. doi: 10.1097/00004703-199906000-00005. [DOI] [PubMed] [Google Scholar]
  37. Liew J, Eisenberg N, Spinrad TL, Eggum ND, Haugen RG, Kupfer A, … Baham ME. Physiological regulation and fearfulness as predictors of young children’s empathy-related reactions: Temperament and empathy. Social Development. 2011;20:111–134. doi: 10.1111/j.1467-9507.2010.00575.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Monroe SM, Simons AD. Diathesis-stress theories in the context of life stress research: Implications for depressive disorders. Psychological Bulletin. 1991;110:406–425. doi: 10.1037/0033-2909.110.3.406. [DOI] [PubMed] [Google Scholar]
  39. Pesonen AK, Räikkönen K, Paavonen EJ, Heinonen K, Komsi N, Lahti J, … Strandberg T. Sleep duration and regularity are associated with behavioral problems in 8-year-old children. International Journal of Behavioral Medicine. 2010;17:298–305. doi: 10.1007/s12529-009-9065-1. [DOI] [PubMed] [Google Scholar]
  40. Porges SW. The polyvagal perspective. Biological Psychology. 2007;74:116–143. doi: 10.1016/j.biopsycho.2006.06.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Porges SW, Doussard-Roosevelt JA, Portales AL, Greenspan SI. Infant regulation of the vagal “brake” predicts child behavior problems: A psychobiological model of social behavior. Developmental Psychobiology. 1996;29:697–712. doi: 10.1002/(SICI)1098-2302(199612)29:8<697::AID-DEV5>3.0.CO;2-O. [DOI] [PubMed] [Google Scholar]
  42. Roisman GI, Newman DA, Fraley RC, Haltigan JD, Groh AM, Haydon KC. Distinguishing differential susceptibility from diathesis-stress: Recommendations for evaluating interaction effects. Development and Psychopathology. 2012;24:389–409. doi: 10.1017/S0954579412000065. [DOI] [PubMed] [Google Scholar]
  43. Sadeh A. A brief screening questionnaire for infant sleep problems: Validation and findings for an internet sample. Pediatrics. 2004;113:e570–e577. doi: 10.1542/peds.113.6.e570. [DOI] [PubMed] [Google Scholar]
  44. Sadeh A. Consequences of sleep loss or sleep disruption in children. Sleep Medicine Clinics. 2:513–520. doi: 10.1016/j.jsmc.2007.05.012. [DOI] [Google Scholar]
  45. Sadeh A, Raviv A, Gruber R. Sleep patterns and sleep disruptions in school-age children. Developmental Psychology. 2000;36:291–301. doi: 10.1037/0012-1649.36.3.291. [DOI] [PubMed] [Google Scholar]
  46. Staples AD, Bates JE. Children’s sleep deficits and cognitive and behavioral adjustment. In: El-Sheikh M, editor. Sleep and development: Familial and socio-cultural considerations. Oxford University Press; New York, NY: 2011. pp. 133–164. [Google Scholar]
  47. Sulik MJ, Eisenberg N, Spinrad TL, Silva KM. Associations between respiratory sinus arrhythmia (RSA) reactivity and effortful control in preschool-age children. Developmental Psychobiology. 2015;57:596–606. doi: 10.1002/dev.21315. [DOI] [PubMed] [Google Scholar]
  48. Tabachnick BG, Fidell LS. Using multivariate statistics. 5. Boston, MA: Allyn & Bacon/Pearson Education; 2007. [Google Scholar]
  49. Tikotzky L, Sadeh A. Sleep patterns and sleep disruptions in kindergarten children. Journal of Clinical Child Psychology. 2001;30:581–591. doi: 10.1207/S15374424JCCP3004_13. [DOI] [PubMed] [Google Scholar]
  50. Touchette E, Côté SM, Petit D, Liu X, Boivin M, Falissard B, … Montplaisir JY. Short nighttime sleep-duration and hyperactivity trajectories in early childhood. Pediatrics. 2009;124:e985–e993. doi: 10.1542/peds.2008-2005. [DOI] [PubMed] [Google Scholar]
  51. Touchette E, Petit D, Séguin JR, Boivin M, Tremblay RE, Montplaisir JY. Associations between sleep duration patterns and behavioral/cognitive functioning at school entry. Sleep. 2007;30:1213–1219. doi: 10.1093/sleep/30.9.1213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. van Buuren S, Groothuis-Oudshoorn K. MICE: Multivariate imputation by chained equations in R. Journal of Statistical Software. 2011;45:1–67. [Google Scholar]
  53. Willoughby M, Kupersmidt J, Voegler-Lee M, Bryant D. Contributions of hot and cool self-regulation to preschool disruptive behavior and academic achievement. Developmental Neuropsychology. 2011;36:162–180. doi: 10.1080/87565641.2010.549980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Zuckerman M. Sensation seeking: Beyond the optimum level of arousal. Hillsdale, NJ: Erlbaum; 1979. [Google Scholar]

RESOURCES