Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Feb 1.
Published in final edited form as: J Sleep Res. 2016 Feb;25(1):62–69. doi: 10.1111/jsr.12338

Daily Affective Experiences Predict Objective Sleep Outcomes among Adolescents

Royette Tavernier a, Sungsub B Choo a, Kathryn Grant b, Emma K Adam a
PMCID: PMC4791211  NIHMSID: NIHMS714562  PMID: 26365539

Summary

Adolescence is a sensitive period for changes in both sleep and affect. Although past research has assessed the association between affect and sleep among adolescents, few studies have examined both trait (typical) and day-to-day changes in affect, and fewer still have specifically examined negative social evaluative emotions (NSEE; e.g., embarrassment) in relation to sleep. We examined both between- and within-person variations in daily affect in relation to four objectively-measured sleep outcomes (sleep hours, sleep latency, sleep efficiency, and length of wake bouts) among adolescents. Participants (N = 77 high school students, 42.9% female; M = 14.37 years) wore an actiwatch and completed daily diaries for 3 days. Results of hierarchical linear models (controlling for age, gender, race, ethnicity, parental employment status, income, puberty, and caffeine) indicated that NSEE and high arousal affective experiences generally predicted poor sleep outcomes, whereas low arousal affective experiences were associated with good sleep outcomes. Specifically, at the person level, adolescents reporting higher NSEE had shorter average sleep hours, and those experiencing higher anxiety-nervousness had longer wake bouts. In addition, individuals experiencing more dysphoria (sad, depressed, lonely) had longer average sleep hours and shorter wake bouts, while those experiencing more calmness had shorter sleep latencies. At the within person level, individuals had longer sleep latencies following days that they had experienced high arousal positive affect (e.g., excitement) and had longer wake bouts following days they had experienced more NSEE. Results highlight the detrimental effects of NSEE and high arousal affective states for adolescent sleep.

Keywords: actigraphy, daily-diary study, high arousal, low arousal, negative social evaluative emotions

Introduction

Adolescence is a sensitive period for changes in sleep behaviors (Carskadon, 2011). Adolescents report shorter sleep hours, later bedtimes, and more irregular sleep-wake timing compared to children (Wolfson & Carskadon, 1998). These changes in sleep-wake patterns during adolescence are thought to be due to biological changes associated with the onset of puberty, that result in a shift towards a later circadian timing around age 13 years (Spear, 2000; Tonetti, Fabbri, & Natale, 2008). Additionally, social factors such as media use, work and extracurricular activities, increased academic workload, and early class start times also contribute to changes in adolescent sleep-wake patterns (Carskadon, 2011). In addition to being a sensitive period for sleep, adolescence also is characterized by important developmental changes in affective processes (Birditt & Fingerman, 2003). Findings from neuroscience studies, for example, show greater brain activation in response to emotionally-laden content among adolescents relative to adults (Monk et al., 2003). Within the literature, there is overwhelming evidence to support heightened emotional arousal and reactivity among adolescents relative to children and adults (Spear, 2000). Importantly, affect regulation has been shown to have significant implications for various aspects of psychosocial functioning, based on both concurrent and longitudinal findings (e.g., Bandura et al., 2003), and thus remains an important aspect of overall well-being, particularly during the sensitive age period of adolescence.

The recently published sleep recommendations from the U.S National Sleep Foundation indicate that adolescents (14- to 17- year olds) need 8 – 10 hours of sleep per night (National Sleep Foundation, 2015). However, findings consistently show that adolescents from across the globe generally do not meet the minimal recommended sleep hours (Gradisar, Gardner, & Dohnt, 2011). Past research based on both experimental and correlational studies show a consistent significant link between sleep and affect across the lifespan. Among a sample of 11–14 year olds, shorter sleep hours were significantly correlated with higher levels of depressive symptoms (Fredriksen et al., 2004) – a finding that has been confirmed among emerging adults and older adults. Short sleep hours also have been significantly linked to greater feelings of anxiety (Liu & Zhou, 2002). Moreover, Winsler and colleagues (2014) recently found that adolescents who reported getting one hour less sleep during the weekday were at a significantly greater risk for reporting increased feelings of hopelessness and suicidal ideation. Similarly, subjective reports of poor sleep quality (i.e., difficulty initiating or maintaining sleep, degree of restfulness upon waking, and night awakenings etc.) also have been significantly linked to higher scores on negative affect, depressive symptoms, anxiety, and perceived stress (Baglioni et al., 2010). Moreover, findings from experimental studies indicate that sleep restriction leads to more impulsive behaviors among emerging adults, particularly in response to negative stimuli (Anderson & Platten, 2011). Sleep restriction also has been shown to increase one’s propensity to make more risky decisions, as evidenced by higher cost-benefit ratio scores in the context of a lab-based computerized risk-taking task (Rossa et al., 2014). Of note, good sleep also has been linked to non-clinical forms of affect, including positive affect states such as feeling happy, joyful, and cheerful (van Zundert et al., 2015). Thus, given the importance of adequate sleep for optimal psychosocial functioning, particularly during sensitive periods of development, understanding the factors that predict poor sleep among adolescents remains a critical area of research for developmental psychologists.

Although experimental studies of sleep deprivation and its resulting negative influence on emotional functioning provide critical information on the importance of adequate sleep for emotional well-being, these studies do not provide a test of the alternative pathway (i.e., the effect of emotional functioning on sleep outcomes). Importantly, some evidence suggests that the nature of the association between sleep and emotional well-being is bidirectional (Baglioni et al., 2010; Kouros and El-Sheikh 2015; Vandekerckhove & Cluydts 2010). We, therefore, add to the growing body of literature that examine the alternative pathway of the predictive effect of emotional functioning on sleep hours. Furthermore, given the multi-faceted nature of sleep, researchers have found it necessary to incorporate additional sleep parameters beyond sleep hours, such as sleep efficiency, sleep latency, and number of awakenings – all of which have been linked to psychosocial functioning (Cousins et al., 2011; Kouros & El-Sheikh, 2015). In the present study, we assess four sleep parameters (sleep hours, sleep latency, sleep efficiency, and length of wake bouts), which allow us to examine how affect experiences throughout the day are implicated in different aspects of adolescent sleep behaviors at night.

The majority of past studies examining sleep and negative affect have examined between-person differences in sleep, either in experimental or correlational studies (Anderson & Platten, 2011; Winsler et al., 2014). These approaches do not examine within-person variations in affect in relation to sleep. To address this gap, researchers have conducted multiple assessments of sleep and affect across a number of days in order to examine day-to-day or within-person variations in sleep and affect. Kalmbach and colleagues (2014), for example, measured sleep and affect across 14 days and found that higher than average negative affect during the day (for a given individual) was significantly associated with subjective/self-reports of both shorter sleep hours and poorer sleep quality that night. Importantly, some of these studies, specifically, have examined bidirectional associations between sleep and affect. In one recent study, for example, Kouros and El-Sheikh (2015) found that among a community sample of children assessed across 7 consecutive days, longer sleep latency predicted worse next-day mood, and, in turn, worse mood predicted both longer sleep latency and higher sleep activity that night.

The present study extends past research by assessing a broad range of emotions, including negative social evaluative emotions (NSEE). NSEE refer to feelings of rejection, shame, and embarrassment, and thus are rooted in the context of interpersonal relationships (Leary, 2001). Given the importance of peer relationships during adolescence (LaGreca & Harrison, 2005), examining NSEE in relation to sleep may be particularly important during this age period (Dahl & Lewin, 2002). The present study examines both between person differences, as well as daily variations in NSEE. Additionally, the present study examines both high arousal (anxiety-nervousness) and low arousal (dysphoria) negative affective experiences; as well as high arousal (excited, happy, energetic) and low arousal (calm) positive affective experiences in relation to adolescent sleep. Furthermore, although some past studies have examined the link between positive affect and sleep behaviors, the majority of these studies have been conducted with emerging or older adult samples (Kalmbach et al., 2014; McCrae et al., 2008; but see Dagys et al., 2012). To address this gap, the present study examined daily positive affective states in relation to sleep outcomes among a sample of adolescents.

The purpose of the present 3-day daily diary study was to examine the effects of daily affective experiences on four sleep outcomes among a sample of adolescents. We employed hierarchical linear modeling (HLM), which allowed us to examine time-varying (i.e., day-to-day) effects of affective experiences on a particular day on sleep on that particular night, in addition to examining non-time varying (i.e., between-person) differences in sleep outcomes. Although bidirectional associations between sleep and affect are likely to occur, with sleep the night before also predicting affect the next day, due to limited days of data (only 2 days of data in which we had affect data the day after sleep data, as compared to 3 days of data in which we had sleep data after affect data), we focused in the current paper on affective predictors of sleep later that night.

Based on the hyper arousal model of insomnia, which states that physiological and cognitive arousal are key mechanisms that lead to difficulties with initiating and maintaining sleep (Riemann et al., 2010), we hypothesized that low arousal affective states would predict better sleep outcomes, while high arousal affective states would predict worse sleep outcomes, regardless of valence. Additionally, given the importance of interpersonal relationships for adolescents, we hypothesized that NSEE would be particularly strong predictors of poor sleep outcomes.

Method

Participants and Procedures

Participants were a subsample (N = 77) of high school students (11–18 years old) who were part of the larger multi-method study (N = 379) on adolescent adjustment. Participants were recruited from 3 diverse public schools in a Midwestern city in the U.S, and completed measures as part of a full day data collection session at a local university. Measures from the larger study included assessments of chronic and everyday life stressors, emotional functioning, and executive functioning. The participants for the present study were a subsample of adolescents who were invited to participate in the sleep component of the study. Specifically, these participants wore actiwatches and completed paper-copy daily morning (with instructions to complete within 30 minutes of waking) and evening diaries (with instructions to be completed just before bedtime) for 3 days (Sunday through Tuesday). Evening diaries included questions on adolescents’ daily experiences while morning diaries included questions on adolescents’ sleep from the previous night. The study was approved by the Northwestern and DePaul University Ethics boards and participants provided informed consent prior to participation.

Missing Data Analysis

Of the 131 participants (approximately 35% of the larger sample) who were invited to participate in the sleep and daily diary component of the study, 63% had sleep data from actigraphy, 79% returned morning diaries and 89% returned their evening diaries. For the purposes of the present study, we selected only participants who had complete sleep actigraphy data (N = 77). As a result, there were no missing data on the sleep parameters for the participants in our subsample. Of these participants, missing data on the affect variables coded from the evening diaries at Level 1 (i.e., for all data points across 3 days for all 77 participants) ranged from 0% to 16.8%. At Level-2 (i.e., when scores on the affect variables were averaged across the 3 days for each participant), missing data ranged from 0% to 13%. Missing data for the affect variables (and covariates) were estimated using the full information maximum likelihood (FIML) estimation method. FIML retains cases that have missing data, thus avoiding the biased parameter estimates that can occur with pair-wise or list-wise deletion.

Measures

Covariates

Questions on a demographic questionnaire assessed participants’ age; gender (0 = female, 1 = male); race (“Black or African-American”, “Asian or Asian-American”, “American Indian or Alaskan Native”, “Native Hawaiian or Other Pacific Islander”, “White or Caucasian”, “Mixed”; or “Other”); ethnicity (0 = “Not Hispanic or Latino”, 1 = “Hispanic or Latino”); and parental employment status (“Does your mother (father) have a job?”; 0 = employed, 1 = unemployed), while income was assessed from a parent questionnaire. Caffeine use was assessed with the following survey item: “How many caffeinated drinks (coffee or soft drinks with caffeine in them e.g., coke, pepsi or mountain dew) did you have today?” Puberty was assessed with 5 items for females (e.g., growth in height, body hair, skin changes, breast development, and presence of menstrual cycle) and 5 items for males (e.g., growth in height, body hair, skin changes, deepening of voice, and facial hair), based on the Pubertal Development Scale (PDS, Peterson et al., 1988). Response options ranged from 1 = not yet started to 4 = seems completed. Scores were averaged across the 5 items, with higher scores indicating higher pubertal development.

Daily Affective Experiences

As part of their evening diaries, participants were asked to rate the extent to which they had experienced a list of 17 items, based on their overall daily experiences, with response options ranging from 1 = Not at all to 5 = Very much. Results of a principal components analysis with varimax rotation yielded the following 5 multi-item components: i) Negative Social Evaluative Emotions (judged, rejected, embarrassed, ashamed, α = 0.89); ii) Dysphoria (sad, depressed, lonely, α = 0.87); iii) Anxious-nervous, α = 0.82; iv) Angry-irritable, α = 0.77; and v) High Arousal Positive Affect (excited, happy, energetic, α = 0.80). Additionally, we included the following two single items that did not load highly on any one specific component, as separate predictors in the model: vi) Calm; and vii) Tired.

Objective Sleep Measures

Participants wore an Actiwatch Score (Phillips Respironics, Inc.) on their non-dominant hand for 3 days. The Actiwatch is a wrist-band accelerometer that records movement during wakefulness and sleep. The Actiware-Sleep software (version 3.4, MiniMitter/Philips Respironics)-validated algorithm was used to score the data (Oakley, 1997). This algorithm calculates a number of sleep parameters using 1-minute epochs and based on significant movement after at least 10 minutes of inactivity. The following four sleep outcomes were calculated: i) sleep hours: actual amount of time spent asleep, minus periods of awakening; ii) sleep latency: the amount of time between bedtime and sleep onset; iii) sleep efficiency: percentage of actual sleep time as a function of time in bed; and iv) length of wake bouts: average duration of wake bouts between sleep start and sleep end.

Plan of Analysis

In order to account for the nested nature of the data (i.e., days nested within persons), we analyzed a 2-level hierarchical linear model using HLM software (version 6.0). This approach accounts for the fact that the data points are not independent of each other, because we have multiple assessment days per individual (Raudenbush & Bryk, 2002). Level 1 reflects associations between proposed predictor and outcome variables at the day-to-day level within each individual. In other words, at Level 1, we examine how sleep is impacted each night, based on participants’ affective states that day; we are predicting day to day variations in sleep from day to day variations in affect. Level 2, on the other hand, reflects associations between proposed predictor and outcome variables at the person-level. In other words, at Level 2, we examine how average sleep parameters (averaged across the 3 assessment days) are related to individual differences in affective states (also averaged across the 3 assessment days). Using this approach, we predicted the four sleep outcomes from the affect variables (see Measures), both at the day-to-day level (level 1) and the person-level (level 2). Level 1 variables were group-mean centered and level 2 variables were grand-mean centered. The following dummy-coded variables were uncentered, at Level-2: gender, race, ethnicity, and parental employment status. In all four regression models (one for each sleep outcome), we included: age (grand-mean centered), gender, race, ethnicity, parental employment status, caffeine use, and pubertal development as covariates.

  • Level I – Model:

    Sleep Outcometi = Π0i + Π1i *(Caffeineti) + Π2i *(NSEEti) + Π3ti *(Dysphoricti) + Π4i *(HAPAti) + Π5i *(Calmti) + Π6i *(Anxious-Nervousti) + Π7i *(Angry-Irritableti) + Π8i *(Tiredti) + eti

  • Level 2 – Model:

    Π0i = β00 + β01* (Age) + β02* (Gender) + β03* (Black) + β04* (Mixed) + β05* (Other) + β06* (Hispanic) + β07* (Mom not employed) + β08* (Dad not employed) + β09* (Income) + β10* (Caffeine) + β11* (Pubertal Development) + β12* (NSEE) + β13* (Dysphoric) + β14* (HAPA) + β15* (Calm) + β16* (Anxious-Nervous) + β17* (Angry-Irritable) + β18* (Tired) + r0i

Results

Descriptives

Descriptive statistics for all study variables are presented in Table 1. Participants (n = 77) ranged in age from 11 to 18 years (M = 14.37, SD = 1.95). African-American/Black (35%) and White (20.5%) were the two most prevalent racial groups, and the majority (63%) indicated being of Non-Hispanic ethnicity. Average pubertal development was significantly higher for females (M = 3.16, SD = 0.68) relative to males (M = 2.60, SD = 0.81), t(75) = 3.190, p = .002. Overall, actigraph- recorded average bedtimes and wake times were 11:00pm and 7:00am, respectively. Adolescents obtained an average of 6 hrs, 6 minutes sleep; average sleep latency was approximately 23 minutes, while average length of wake bouts was just under 2 minutes, and average sleep efficiency was approximately 81%. Results of a MANOVA indicated no significant gender differences on the sleep outcomes, λ =.998, F (4, 72) = 0.031, p = .998.

Table 1.

Means, Standard Deviations, and Ranges for all Study Variables

Variables Mean (SD) Range
Age 14.37 (1.95) years 11 – 18 years
Gender 50.8% female n/a
Mom employed 79.3% n/a
Dad employed 81.7% n/a
Income 2.70 (1.25) 1 – 6
Race Black (35%), White (20.5%), Mixed (17.1%), Other (27.4%) n/a
Ethnicity Non-Hispanic (63%) n/a
Caffeine 0.86 (1.08) 0.00 – 5.00
Puberty 2.84 (0.80) 1.00 – 4.00
Dysphoria 1.50 (0.63) 1.00 – 5.00
Angry-Irritable 1.75 (0.73) 1.00 – 5.00
Anxious-Nervous 1.62 (0.68) 1.00 – 5.00
Tired 2.92 (0.99) 1.00 – 5.00
Negative Social Evaluative Emotions 1.25 (0.46) 1.00 – 5.00
High Arousal Positive Affect 2.60 (0.77) 1.00 – 5.00
Sleep hours 6.10 (0.96) 3.98 – 8.12
Sleep latency (minutes) 23.40 (24.60) 0.00 – 109.20
Sleep efficiency (%) 80.99 (7.39) 60.90 – 91.83
Length of wake bouts (minutes) 1.80 (0.60) 1.20 – 4.20
Open in a new tab

Note. N = 77

Sleep Hours

Table 2 shows the results of the predictive effects of both person-level and day-to-day affective experiences for each of the four sleep outcomes. At the person-level, higher NSEE predicted shorter sleep hours, B = −0.372, SE = 0.129, p = 0.005; whereas higher dysphoria predicted longer sleep hours, B = 0.327, SE = 0.157, p =0.041. At the person-level, higher levels of calmness was associated with longer sleep hours at the trend level, B = 0.214, SE = 0.121, p = 0.081.

Table 2.

Results of Hierarchical Linear Models of Between and Within Daily Affect Predictors of Sleep Outcomes

Sleep Hours Sleep Latency Sleep Efficiency Average Length of Wake Bouts
Fixed Effect Coefficient (SE) Coefficient (SE) Coefficient (SE) Coefficient (SE)
Intercept1, π0
Intercept2, β00 0.303 (0.203) −0.422 (0.155) 0.560 (0.220) −0.261 (0.249)
Person-Level Predictors
Negative Social Evaluative Emotions, β012 −0.372 (0.129)** 0.231 (0.145) −0.134 (0.168) 0.085 (0.157)
Dysphoric, β013 0.327 (0.157)* −0.085 (0.130) 0.221 (0.147) −0.335 (0.123)**
High Arousal Positive Affect, β014 −0.004 (0.117) 0.380 (0.161)* 0.163 (0.001) 0.139 (0.124)
Calm, β015 0.214 (0.121)+ −0.440 (0.138)** 0.260 (0.141)+ 0.059 (0.114)
Anxious-Nervous, β016 −0.020 (0.110) −0.193 (0.136) 0.152 (0.163) 0.357 (0.150)*
Angry-Irritable, β017 −0.106 (0.122) −0.135 (0.174) −0.057 (0.196) 0.249 (0.172)
Tired, β018 −0.111(0.136) 0.136 (0.165) −0.040 (0.189) −0.057 (0.122)
Within-Person Predictors
Negative Social Evaluative Emotions, β20 0.014 (0.151) 0.146 (0.165) −0.157 (0.132) 0.302 (0.141)*
Dysphoric, β30 −0.040 (0.094) 0.078 (0.132) 0.029 (0.105) −0.056 (0.160)
High Arousal Positive Affect, β40 0.005 (0.113) 0.248 (0.120)* −0.179 (0.105)+ 0.021 (0.184)
Calm, β50 0.061 (0.081) −0.026 (0.120) 0.054 (0.104) −0.181 (.109)+
Anxious-Nervous, β60 0.119 (0.110) −0.106 (0.162) 0.126 (0.133) 0.198 (0.149)
Angry-Irritable, β70 −0.064 (0.086) −0.036 (0.108) −0.050 (0.092) −0.105 (0.101)
Tired, β80 0.041 (0.112) 0.155 (0.088)+ −0.047 (0.075) 0.020 (0.132)
Open in a new tab

Note.

*

p < .05,

**

p < .01,

***

p < .001,

+

p < .10

Sleep Latency

At the person-level, higher feelings of calmness predicted shorter sleep latency B = −0.440, SE = 0.138, p = 0.002; whereas high arousal positive affect predicted longer sleep latency both at the person-level, B = −0.380, SE = 0.161, p = 0.022, and at the day-level, B = 0.248, SE = 0.120, p = 0.041. At the within-person level, there was a trend for longer sleep latencies on days when adolescents reported feeling more tired, B = 0.155, SE = 0.088, p = 0.081.

Sleep Efficiency

At the person-level, there was a trend-level finding, such that higher feelings of calmness predicted higher sleep efficiency, (B = 0.260, SE = 0.141, p = 0.071). There was also a trend for lower sleep efficiency following days when adolescents reported higher high arousal positive affect, (B = 0.179, SE = 0.105, p = 0.092).

Length of Wake Bouts

At the person-level, higher feelings of anxious-nervousness predicted longer wake bouts, B = 0.357, SE = 0.150, p = 0.020; whereas higher feelings of dysphoria predicted shorter wake bouts, B = −0.335, SE = 0.123, p = 0.008. At the within-person level, higher NSEE predicted longer wake bouts, B = 0.302, SE = 0.141, p = 0.034. At the within-person level, there was a trend-level finding for shorter wake bouts following days when adolescents reported higher feelings of calmness, (B = −0.181, SE = 0.109, p = 0.099).

Discussion

The purpose of the present study was to examine the predictive effects of daily affect experiences on four objectively-measured sleep outcomes among a sample of adolescents. Results indicated that NSEE had negative implications for adolescents’ sleep hours and duration of wake bouts. Feelings of low arousal affective experiences – regardless of valence – were associated with more favourable sleep outcomes, while high arousal affective experiences – regardless of valence – were associated with less favourable sleep outcomes. Results indicated that adolescents’ NSEE predicted poorer sleep outcomes both at the between-person (shorter sleep hours) and within-person-levels (longer wake bouts). Adolescence is a sensitive period for the development of peer relationships and such relationships are crucial to adolescent identity development and psychosocial functioning (Blakemore, 2008). Results of the present study showed that when adolescents’ social relationships give way to feelings of perceived social judgement, the experience negatively impacts sleep behaviors. Among their sample of 7–14 year olds, Alfano and colleagues (2010) found that cognitive arousal (e.g., “I think about things that happened during the day”) prior to sleep onset was significantly correlated with disturbed nocturnal sleep. Thus, increased levels of cognitive arousal may be one mechanism linking NSEE and poor sleep. Ruminating over negative aspects of social experiences also may explain the link between NSEE and poor sleep (Thomsen et al., 2003). There is some evidence that more interpersonal difficulties with peers are significantly associated with poor subjective sleep quality (Sarchiapone et al., 2014). Negative social-evaluative emotions have also been linked to elevations in the stress hormone cortisol (Dickerson et al., 2004), and elevated evening cortisol, in turn has been associated with shorter sleep hours (Zeiders et al., 2011), suggesting one possible biological pathway by which NSEE are associated with shorter sleep. The present study, extends past research by suggesting that NSEE may be particularly important aspects of adolescents’ emotional experience to consider in relation to objective sleep outcomes.

Moreover, as hypothesized, low arousal affective experiences predicted more favourable sleep outcomes. Specifically, dysphoria predicted longer sleep hours and shorter wake bouts at the person-level. Feeling calm predicted shorter sleep latency (and was associated with higher sleep efficiency, longer sleep hours, and shorter wake bouts at the trend-level). These two predictors (dysphoria and calm), however, being opposite in valence may be associated with sleep outcomes in different ways. For example, the significant link between dysphoria and longer sleep hours may be due to the use of sleep as a coping mechanism among those who are overwhelmed by sadness, depressed mood, and loneliness. Indeed, long sleep has been noted as a marker of depression, and also may be a side effect of anti-depressant medications (Thase, 2006). Importantly, findings indicated that feeling calm was protective for adolescent sleep (e.g., shorter sleep latency). As a low-arousal positive affective state, feeling calm may protect against cognitive intrusions, which have been shown to interfere with sleep initiation and maintenance (Alfano et al., 2010). Similarly, Howell and colleagues (2008) found that mindfulness was significantly associated with higher subjective reports of good sleep among the sample of young adults. Future research should examine the specific sources of calmness among adolescents to better delineate the ways in which school administrators and parents can support adolescents in experiencing increased feelings of calmness.

Furthermore, results of within-person analyses indicated that adolescents reported longer sleep latency on days when they reported higher than average high arousal positive affect (HAPA). Additionally, adolescents who typically report experiencing higher levels of high arousal positive affect also reported longer sleep latency across the three days (person-level). There was also a trend-level findings for lower sleep efficiency on days when adolescents reported higher high arousal positive affect states. The fact that excitement makes it more difficult for adolescents to fall asleep should perhaps not be surprising (given the overwhelming empirical support for heightened arousal and poor sleep; Riemann et al., 2010), but associations between adolescent positive emotion and objective sleep have not been extensively explored in past research on adolescent emotion and sleep, which has tended to focus on negative affective states (e.g., Winsler et al., 2014). Not surprisingly, feelings of anxious-nervousness predicted longer wake bouts (index of difficulty maintaining sleep) at the person-level. The physiological implications of these highly arousing affective states (regardless of valence) may interfere with adolescents’ ability to initiate and maintain sleep (Cohen & Pressman, 2006). Increased cognitive arousal also may help explain these findings (Alfano et al., 2010).

Results of the present study should be interpreted in light of the study’s limitations. First, participants in the present study completed daily diary measures and wore an actiwatch for only 3 days. Future research should test the model in the present study with a sample of adolescents assessed across a longer number of days (including weekends) in order to draw comparisons between the effects of the predictors on school-night versus weekend sleep. Secondly, we note that, in terms of participant compliance to diary reports, there was no objective way to determine whether participants completed their diaries daily, or retroactively completed diaries for a string of days simultaneously. However, all participants received a Participant Instruction Booklet with detailed information about the study’s procedures, with emphasis on the importance of adhering to the study’s daily protocol. Additionally, no additional compensation was given for higher diary completion rates, thus possibly removing any incentive to retroactively complete multiple diaries simultaneously. Nevertheless, we suggest that future studies should employ an objective measure (e.g., automated dated stamp) for determining participant compliance with completion of daily diaries.

Third, given the correlational nature of the data, our findings cannot speak to any causal associations between variables. The nature of the associations between affective experiences and sleep variables are likely bidirectional (Baglioni et al., 2010). In fact, empirical evidence from experimental studies show that sleep restriction/deprivation negatively impacts affect/emotion reactivity (e.g., Baum et al., 2014), perhaps due to reduced connectivity between the medial-prefrontal cortex and the amygdala (Yoo et al., 2007). Furthermore, within the context of daily diary studies, evidence indicates significant bidirectional associations between poor sleep and negative emotional functioning (Kouros & El-Sheikh, 2015). Thus, one limitation of the present study was the lack of assessment of bidirectional associations between the proposed affect predictors and sleep outcomes. The lack of assessment of bidirectional associations in the present study was due to the limited number of assessment days and relatively small sample size. We propose, therefore, that future research should include more days of testing in order to assess the predictive effect of objectively measured sleep parameters on a range of next day affective states (ranging in both arousal and valence), in addition to assessing how these various affective states throughout the day impact same day sleep. Moreover, it will be important for future studies to specifically assess the timing of adolescents’ affective experiences and examine how this relates to sleep behaviors. For example, are high arousal affective experiences more detrimental to adolescent sleep if experienced later in the day (i.e., closer to bedtime) compared to earlier in the day? Relatedly, future research based long-term longitudinal designs are further needed to specifically test for mediating pathways that account for the significant effects found between the various affective states and sleep parameters in the present study. Fourth, given the homogeneity of our sample, findings may not be generalizable to other populations such as emerging or older adults, or individuals with clinical levels of sleep and mood disorders.

Despite these limitations, the present study makes a number of important contributions to past research. Importantly, we incorporated the use of daily assessments of various affective experiences and sleep outcomes. This approach provides insight into how day-to-day variability in adolescents’ everyday affective experiences relate to daily variability in sleep (within-person), as well as how more relatively stable individual differences in affective experiences relate to individual differences in sleep (between-person). In terms of our predictors, we included assessments of both high arousal and low arousal negative and positive affective experiences, and specifically examined the role of negative social evaluative emotions in adolescent sleep. Our findings highlight the need for researchers to further investigate the mechanisms through which NSEE negatively impact adolescent sleep. We found evidence that highly arousing affective states (regardless of valence) generally had detrimental effects on adolescent sleep, while low arousing affective states were linked to better sleep outcomes. Results of the present study can be used to educate adolescents, parents, and school administration about the different aspects of adolescents’ daily experiences that have significant implications for sleep behaviors, in hopes of improving adolescent sleep. Given the crucial role that sleep plays for optimal psychosocial functioning across the lifespan, researchers should continue investigating sources of variability in adolescents’ affective experiences and further examine mechanisms that explain the associations among affective experiences and sleep during adolescence.

Acknowledgments

This research was supported by NIAAA R21 HL109449-01.

Footnotes

Royette Tavernier was involved in the conception of the present study, analyzed the data, and drafted the manuscript. Sungsub B Choo conducted preliminary analyses of the data. Kathryn Grant was involved in the design of the larger study and collection of data. Emma K Adam was involved in the conception of the study, collection of data, analyzed the data and edited the manuscript.

References

  1. Alfano CA, Pina AA, Zerr AA, Villalta IK. Pre-sleep arousal and sleep problems of anxiety-disordered youth. Child Psychiatry Hum Dev. 2010;41:156–167. doi: 10.1007/s10578-009-0158-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anderson C, Platten CR. Sleep deprivation lowers inhibition and enhances impulsivity to negative stimuli. J Behav Brain Sci. 2011;217:463–466. doi: 10.1016/j.bbr.2010.09.020. [DOI] [PubMed] [Google Scholar]
  3. Baglioni C, Spiegelhalder K, Lombardo C, Riemann D. Sleep and emotions: a focus on insomnia. Sleep Med Rev. 2010;14:227–238. doi: 10.1016/j.smrv.2009.10.007. [DOI] [PubMed] [Google Scholar]
  4. Bandura A, Caprara GV, Barbaranelli C, Gerbino M, Pastorelli C. Role of affective self-regulatory efficacy in diverse spheres of psychosocial functioning. Child Dev. 2003;74:769–782. doi: 10.1111/1467-8624.00567. [DOI] [PubMed] [Google Scholar]
  5. Baum KT, Desai A, Field J, Miller LE, Rausch J, Beebe DW. Sleep restriction worsens mood and emotion regulation in adolescents. J Child Psychol Psychiatr. 2014;55:180–190. doi: 10.1111/jcpp.12125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Birditt KS, Fingerman KL. Age and gender differences in adults’ descriptions of emotional reactions to interpersonal problems. J Gerontol B Psychol Sci Soc Sci. 2003;58:237–245. doi: 10.1093/geronb/58.4.p237. [DOI] [PubMed] [Google Scholar]
  7. Blakemore SJ. The social brain in adolescence. Nat Rev Neurosci. 2008;9:267–277. doi: 10.1038/nrn2353. [DOI] [PubMed] [Google Scholar]
  8. Bryk AS, Raudenbush SW. Hierarchical linear models: applications and data analysis methods. 2 Sage Publications, Inc; 1992. [Google Scholar]
  9. Carskadon MA. Sleep in adolescents: the perfect storm. Pediatr Clin North Am. 2011;58:637–647. doi: 10.1016/j.pcl.2011.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cohen S, Pressman SD. Positive affect and health. Curr Dir Psychol Sci. 2006;15:122–125. [Google Scholar]
  11. Cousins JC, Whalen DJ, Dahl RE, Forbes EE, Olino TM, Ryan ND, Silk JS. The bidirectional association between daytime affect and nighttime sleep in youth with anxiety and depression. J Pediatr Psychol. 2011;36:969–979. doi: 10.1093/jpepsy/jsr036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dagys N, McGlinchey EL, Talbot LS, Kaplan KA, Dahl RE, Harvey AG. Double trouble? The effects of sleep deprivation and chronotype on adolescent affect. J Child Psychol Psychiatry. 2012;53:660–667. doi: 10.1111/j.1469-7610.2011.02502.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dahl RE, Lewin DS. Pathways to adolescent health sleep regulation and behavior. J Adolesc Health. 2002;31:175–184. doi: 10.1016/s1054-139x(02)00506-2. [DOI] [PubMed] [Google Scholar]
  14. Dickerson Sally S, Gruenewald Tara L, Kemeny Margaret E. When the social self is threatened: Shame, physiology, and health. J Pers. 2004;72:1191–1216. doi: 10.1111/j.1467-6494.2004.00295.x. [DOI] [PubMed] [Google Scholar]
  15. El-Sheikh M, Buckhalt JA, Keller PS, Granger DA. Children’s objective and subjective sleep disruptions: links with afternoon cortisol levels. Health Psychol. 2008;27:26–33. doi: 10.1037/0278-6133.27.1.26. [DOI] [PubMed] [Google Scholar]
  16. Folkman S, Lazarus RS. The relationship between coping and emotion: Implications for theory and research. Soc Sci Med. 1988;26:309–317. doi: 10.1016/0277-9536(88)90395-4. [DOI] [PubMed] [Google Scholar]
  17. Fredriksen K, Rhodes J, Reddy R, Way N. Sleepless in Chicago: tracking the effects of adolescent sleep loss during the middle school years. Child Dev. 2004;75:84–95. doi: 10.1111/j.1467-8624.2004.00655.x. [DOI] [PubMed] [Google Scholar]
  18. Gradisar M, Gardner G, Dohnt H. Recent worldwide sleep patterns and problems during adolescence: a review and meta-analysis of age, region, and sleep. Sleep Med. 2011;12:110–118. doi: 10.1016/j.sleep.2010.11.008. [DOI] [PubMed] [Google Scholar]
  19. Hamilton NA, Nelson CA, Stevens N, Kitzman H. Sleep and psychological well-being. Social Indic Res. 2007;82:147–163. [Google Scholar]
  20. Howell AJ, Digdon NL, Buro K, Sheptycki AR. Relations among mindfulness, well-being, and sleep. Pers Individ Dif. 2008;45:773–777. [Google Scholar]
  21. Jackowska M, Dockray S, Hendrickx H, Steptoe A. Psychosocial factors and sleep efficiency: discrepancies between subjective and objective evaluations of sleep. Psychosom Med. 2011;73:810–816. doi: 10.1097/PSY.0b013e3182359e77. [DOI] [PubMed] [Google Scholar]
  22. Kalmbach DA, Pillai V, Roth T, Drake CL. The interplay between daily affect and sleep: a 2-week study of young women. J Sleep Res. 2014;23:636–645. doi: 10.1111/jsr.12190. [DOI] [PubMed] [Google Scholar]
  23. Kelly RJ, El-Sheikh M. Reciprocal relations between children’s sleep and their adjustment over time. Dev Psychol. 2014;50:1137–1147. doi: 10.1037/a0034501. [DOI] [PubMed] [Google Scholar]
  24. Kouros CD, El-Sheikh M. Daily mood and sleep: reciprocal relations and links with adjustment problems. J Sleep Res. 2015;24:24–31. doi: 10.1111/jsr.12226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. La Greca AM, Harrison HM. Adolescent peer relations, friendships, and romantic relationships: Do they predict social anxiety and depression? J Clin Child Adolesc Psychol. 2005;34:49–61. doi: 10.1207/s15374424jccp3401_5. [DOI] [PubMed] [Google Scholar]
  26. Leary MR. Affect, Cognition, and the Social Emotions. Feeling and thinking: The role of affect in social cognition. 2001:331. [Google Scholar]
  27. Liu X, Zhou H. Sleep duration, insomnia and behavioral problems among Chinese adolescents. Psychiatry Res. 2002;111:75–85. doi: 10.1016/s0165-1781(02)00131-2. [DOI] [PubMed] [Google Scholar]
  28. McCrae CS, McNamara JP, Rowe MA, Dzierzewski JM, Dirk J, Marsiske M, Craggs JG. Sleep and affect in older adults: using multilevel modeling to examine daily associations. J Sleep Res. 2008;17:42–53. doi: 10.1111/j.1365-2869.2008.00621.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Monk CS, McClure EB, Nelson EE, Zarahn E, Bilder RM, Leibenluft E, Charney DS, Ernst M, Pine DS. Adolescent immaturity in attention-related brain engagement to emotional facial expressions. Neuroimage. 2003;20:420–428. doi: 10.1016/s1053-8119(03)00355-0. [DOI] [PubMed] [Google Scholar]
  30. Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, Nissen C. The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev. 2010;14:19–31. doi: 10.1016/j.smrv.2009.04.002. [DOI] [PubMed] [Google Scholar]
  31. Rossa KR, Smith SS, Allan AC, Sullivan KA. The effects of sleep restriction on executive inhibitory control and affect in young adults. J Adolesc Health. 2014;55:287–292. doi: 10.1016/j.jadohealth.2013.12.034. [DOI] [PubMed] [Google Scholar]
  32. Sarchiapone M, Mandelli L, Carli V, Iosue M, Wasserman C, Hadlaczky G, Wasserman D. Hours of sleep in adolescents and its association with anxiety, emotional concerns, and suicidal ideation. Sleep Med. 2014;15:248–254. doi: 10.1016/j.sleep.2013.11.780. [DOI] [PubMed] [Google Scholar]
  33. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Methods. 2002;7:147. [PubMed] [Google Scholar]
  34. Silk JS, Steinberg L, Morris AS. Adolescents’ emotion regulation in daily life: Links to depressive symptoms and problem behavior. Child Dev. 2003;74:1869–1880. doi: 10.1046/j.1467-8624.2003.00643.x. [DOI] [PubMed] [Google Scholar]
  35. Spear LP. The adolescent brain and age-related behavioral manifestations. Neurosci Biobehav Rev. 2000;24:417–463. doi: 10.1016/s0149-7634(00)00014-2. [DOI] [PubMed] [Google Scholar]
  36. Thase ME. Depression and sleep: pathophysiology and treatment. Dialogues Clin Neurosci. 2006;8:217–226. doi: 10.31887/DCNS.2006.8.2/mthase. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Thomsen DK, Mehlsen MY, Christensen S, Zachariae R. Rumination—relationship with negative mood and sleep quality. Pers Individ Dif. 2003;34:1293–1301. [Google Scholar]
  38. Tonetti L, Fabbri M, Natale V. Sex difference in sleep-time preference and sleep need: a cross-sectional survey among Italian pre-adolescents, adolescents, and adults. Chronobiol Int. 2008;25:745–759. doi: 10.1080/07420520802394191. [DOI] [PubMed] [Google Scholar]
  39. Vandekerckhove M, Cluydts R. The emotional brain and sleep: an intimate relationship. Sleep Med Rev. 2010;14:219–226. doi: 10.1016/j.smrv.2010.01.002. [DOI] [PubMed] [Google Scholar]
  40. van Zundert RM, van Roekel E, Engels RC, Scholte RH. Reciprocal associations between adolescents’ night-time sleep and daytime affect and the role of gender and depressive symptoms. J of Youth Adolesc. 2015;44:556–569. doi: 10.1007/s10964-013-0009-3. [DOI] [PubMed] [Google Scholar]
  41. Winsler A, Deutsch A, Vorona RD, Payne PA, Szklo-Coxe M. Sleepless in Fairfax: The difference one more hour of sleep can make for teen hopelessness, suicidal ideation, and substance use. J Youth Adolesc. 2014:1–17. doi: 10.1007/s10964-014-0170-3. [DOI] [PubMed] [Google Scholar]
  42. Wolfson AR, Carskadon MA. Sleep schedules and daytime functioning in adolescents. Child Dev. 1998;69:875–887. [PubMed] [Google Scholar]
  43. Wolfson AR, Spaulding NL, Dandrow C, Baroni EM. Middle school start times: the importance of a good night’s sleep for young adolescents. Behav Sleep Med. 2007;5:194–209. doi: 10.1080/15402000701263809. [DOI] [PubMed] [Google Scholar]
  44. Yoo SS, Gujar N, Hu P, Jolesz FA, Walker MP. The human emotional brain without sleep—a prefrontal amygdala disconnect. Curr Biol. 2007;17:R877–R878. doi: 10.1016/j.cub.2007.08.007. [DOI] [PubMed] [Google Scholar]
  45. Zeiders KH, Doane LD, Adam EK. Reciprocal relations between objectively measured sleep patterns and diurnal cortisol rhythms in late adolescents. J Adolesc Health. 2011;48:566–571. doi: 10.1016/j.jadohealth.2010.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES