Skip to main content

Children’s Emotional Expressivity After Sleep Restriction Forecasts Social Problems Years Later


Sleep patterns affect children’s socioemotional functioning in ways that may predict long-term social problems. However, precise mechanisms through which these effects occur remain unexplored and thus unknown. Building on findings in adults, the current study examined whether changes in children’s facial expressions of emotion after sleep restriction predict social problems concurrently and/or longitudinally. At time 1, 37 children (mean = 9.08 years, SD = 1.3) completed in-lab emotional assessments both when rested and after two nights of sleep restriction. Participants’ parents provided reports of their child’s social problems at time 1 and approximately 2 years later (time 2; mean = 11.26 years, SD = 1.6). Children who exhibited less positive facial expressions in response to positive images after sleep restriction evidenced greater social problems longitudinally, even when controlling for earlier social problems. Results suggest that inadequate sleep may undermine children’s social functioning via alterations in emotional expression which may become more salient with age.


Early sleep patterns are closely tied to socioemotional development, or a child’s developing capacity to regulate their emotions and behaviors and form healthy relationships. Accordingly, a burgeoning body of research indicates poor sleep quality in childhood forecasts lower social competence and greater peer/social problems concurrently (Foley & Weinraub, 2017; Vaughn et al., 2015) and over time (Brand et al., 2015; Larose-Grégoire et al., 2018). However, the precise mechanisms through which children’s nighttime sleep impacts their daytime social lives remain relatively unknown. Experimental studies find children report decreases in positive affect (Alfano et al., 2020; Reddy et al., 2017) and increases in negative affect (Baum et al., 2014; McMakin et al., 2016) following sleep loss, yet understanding of specific sleep-dependent behavioral changes that might undermine children’s social relationships is lacking. In the current study, we focus on one salient aspect of social communication: emotional expression.

Emotional expression is a central aspect of social communication that holds increasing importance for peer relationships across early development. Facial expressions not only provide others with an understanding of one’s internal emotional state but are known to have a contagion effect on the emotions of others (Hsee et al., 1993). Children continue to learn how to use appropriate facial expressions of emotion well into the late childhood years (Ekman et al., 1980; Gosselin et al., 2011) and less appropriate emotional displays are more common among peer-rejected children (Hubbard, 2001). After sleep deprivation, children and adults show significantly reduced facial expression of emotion, particularly positive emotion (Alfano et al., 2020; Minkel et al., 2011), and both adults and adolescents express fewer vocal expressions of positive emotion (McGlinchey et al., 2011). No study that we are aware of, however, has examined whether changes in emotional expression following sleep restriction relates to children’s social functioning, acutely or over time. For this study, we utilized data from a previous investigation in our lab where we examined the effects of two nights of partial sleep restriction on children’s emotional processing (Alfano et al., 2020). A majority of these children were also assessed approximately 2 years later. We therefore examined whether sleep restriction-induced changes in emotional facial expressions predicted parent-reported social problems concurrently and/or longitudinally. Guided by available research in adults and adolescents, we expected decreases in positive emotional expressions would be associated with greater social problems at both time points.



Participants included 7- to 11-year-old healthy children originally recruited for an experimental sleep restriction study (see Alfano et al., 2020). Child eligibility requirements for the study included: living with a primary caregiver fulltime, fluency in English, and enrollment in regular education classes (e.g., not in special education classrooms). Participants were excluded if they had any chronic medical problems or conditions known to affect sleep, any psychiatric or sleep disorders, a body mass index > 25 (i.e., increased risk of breathing-related sleep disorders), were taking over the counter or prescription medication that might affect sleep, had an IQ < 80, and/or current or previous suicide ideation. The original sample included 53 children who were contacted via phone or email approximately 2 years later (time 2; T2) about participating in a follow-up study. Forty-four (83%) children from the time 1 (T1) sample participated at T2. The current study includes subsample (N = 37) of children who had complete emotional expression data at T1. As described by Alfano et al. (2020), missing facial expression data occurred if the child wore eyeglasses (i.e., the facial recognition software cannot reliably detect emotions in these participants), had dark skin tone, or because of artifact (e.g., if the child turned away from the camera). Children with and without missing emotional expression data were compared in terms of typical sleep patterns during one-week of actigraphy and social problems and anxiety/depression subscale scores on the Child Behavior Checklist at both T1 and T2. No significant differences were found. Our final sample included 37 children at T1 (M age = 9.08 years, SD = 1.31) and 31 children at T2 (M age = 11.26 years, SD = 1.7). Demographic characteristics of the sample at both time points are provided in Table 1.

Table 1 Demographic, socioemotional, and sleep variables at time 1 and time 2


All study procedures were approved by the Institutional Review Board of the University of Houston. At both time points, children (and one parent) completed an in-person assessment to obtain consent/assent, complete structured psychiatric interviews, IQ testing, and a battery of parent and child questionnaires. All assessments were conducted by trained doctoral students or postdoctoral fellows and reviewed with a licensed clinical psychologist. At T1, children underwent a night of home-based polysomnography (PSG) to ensure a full night of sleep prior to the first emotional assessment the next day. Children then wore actigraphs (Micro MotionLogger, Ambulatory Monitoring, Inc.) for nine consecutive nights. On the second to the last night of actigraphy, parents restricted their child’s sleep at home to 7 h (23:00 to 06:00) according to instructions (and verified with actigraphy). On the final night of actigraphy monitoring, children (with a parent) completed a second night of sleep restriction in the lab including PSG monitoring. Children were permitted to sleep for 6 h (0:00 to 06:00) on the second sleep restriction night. The following morning, at the same time as the first assessment, children completed a second emotional assessment. Emotional tasks in the second assessment were identical to the first assessment using matched counterbalanced stimuli. No significant differences in children’s responses were found based on viewing order. See Alfano et al. (2020) for a full description of the original study.

The T2 assessment took place an average of 782 days (2.14 years; SD = 0.60) after the T1 assessment. Participants who provided consent/assent to participate at T2 completed an in-person assessment including a structured clinical interview, child and parent questionnaires, and 1 week of actigraphy monitoring.

Social Problems

Child Behavior Checklist

At both time points, parents completed the Child Behavior Checklist (CBCL; T. Achenbach & Rescorla, 2001), a 113-item, parent-reported measure assessing a broad range of child behaviors, social and academic functioning. Parents rated their children’s current behavior on a 3-point Likert scale (0 = not true, 1 = somewhat or sometimes true, and 2 = very true or often true). For the current study, the social problems and anxiety/depression subscales (T-scores) were used. The CBCL is one of the most extensively used and validated child rating scales available.

Emotional Assessment Task

Affective Images

During both the rested (baseline) and sleep-restricted T1 emotional assessments, participants were presented with a series of computerized positive and negative images from the International Affective Picture System (IAPS; Lang et al., 2005). During the task, images were displayed on the computer monitor for 6 s preceded by a fixation cross for 3 s. Images were matched and displayed in a random counterbalanced fashion across the two assessments based on content and published norms for arousal and valence ratings. See Alfano et al. (2020) for information about specific images.

Facial Expression of Emotion

Emotional Expression

Facial expressions of emotion were recorded and analyzed using FaceReader 4.0 software (Noldus Inc.). Photography grade lighting equipment was used to ensure adequate and consistent lighting conditions across all assessments. FaceReader detects the face using the Viola-Jones algorithm (Viola & Jones, 2001) and models the face using the Active Appearance method (Cootes et al., 1998). Activation values are calculated for every single frame (30 frames per second) for six basic emotions. Analyses for every frame were manually checked for artifact. Summary scores for valence (i.e., pleasantness) and arousal (i.e., intensity) were used in the current study to minimize the number of models examined. The FaceReader program shows excellent reliability with human coding, with agreement rates up to 95%. To account for individual differences in facial features/expression, all participants were individually calibrated during a resting baseline period. For the current study, scores for each frame were averaged for each 6 s image.

Analytic Plan

Normality was first assessed and those variables showing skewness and/or kurtosis were transformed to standardize scores. Because valence and arousal ratings tended to correlate across image types (e.g., valence ratings for positive and negative images were correlated whereas positive arousal and positive valence ratings were not), separate models based on image type were examined to avoid multi-collinearity. Bivariate correlations among child age, CBCL social problems, and facial expressions were examined. Child age was not associated with CBCL social problems at either time point. However, T1 age was associated with valence of facial expressions for negative images and was therefore included as a covariate in concurrent models. Four linear regression models were used to examine whether sleep-restricted changes in valence or arousal of facial expressions in response to negative and positive images predicted parent-reported social problems at T1 and T2. All models controlled for CBCL anxiety/depressive symptoms at the same time point. T1 CBCL social problems scores and time between T1 and T2 were included as a covariate in models predicting T2 social problems.


Bivariate Correlations

A correlation matrix with all study variables is presented in Table 2. Habitual sleep variables at T1 and T2 were non-significantly associated with CBCL anxiety/depression scores and CBCL social problems scores at both time points. Habitual sleep variables at T1 and T2 were also non-significantly associated with changes in valence or arousal of emotional expression after sleep restriction.

Table 2 Bivariate correlations between study variables

Concurrent Associations Between Emotional Expression and Social Problems

Including child age and T1 CBCL anxiety/depression symptoms as covariates, the model examining whether changes in facial expressions (arousal or valence) in response to positive IAPS images after sleep restriction predicted T1 CBCL social problems scores was non-significant. Similarly, the model examining whether changes in facial expressions in response to negative IAPS images predicted T1 social problems was non-significant. See Table 3.

Table 3 Hierarchical regression models predicting child social problems at T1 and T2

Longitudinal Associations Between Emotional Expression and Social Problems

Controlling for T1 CBCL social problems, number of days between T1 and T2 assessments, and T2 CBCL anxiety/depression symptoms, a decrease in valence of facial expressions in response to positive images (i.e., less positive/pleasant expression of emotion) after sleep restriction significantly predicted CBCL social problems at T2. No other facial expression variable was a significant predictor of T2 CBCL social problems. See Table 3 for all models.


Children with poor sleep quality have more negative interactions with peers, show fewer prosocial behaviors, and report lower friendship satisfaction than good sleepers (Brand et al., 2015; Larose-Grégoire et al., 2018; Vaughn et al., 2015), but mechanisms driving these associations are not understood. In the current study, we explored whether healthy children’s emotional expressions after two nights of partial sleep restriction predicted social problems either concurrently or two years later. Although concurrent relationships were non-significant, less positive facial expressions of emotion after sleep restriction predicted significantly greater social problems at follow-up. The fact that we found longitudinal but not concurrent links between sleep-dependent emotional expressions and child social problems may be explained by developmental differences in sleep, emotional processing, and/or peer relationships. Healthy children recruited for the current study largely evidenced adequate amounts of sleep at T1, which may have reduced our ability to detect significant relationships with social problems at this time point. Emotion recognition ability and the capacity to produce intentional, accurate emotional expressions continue to develop through the age of about 9 (Ekman et al., 1980). Facial expressions of emotion, as opposed to more explicit behaviors such as sharing and cooperativeness (Dekovic & Gerris, 1994) may therefore be less salient predictors of social acceptance/liking at younger ages. Additionally, younger children have less stability in and autonomy over their friendships (i.e., greater parental influence and monitoring) which may be protective for problems with peers (Cairns et al., 1995).

Observation that less positive facial expressions in response to affective images predicted subsequent social problems aligns with research in both children and adults showing positive emotional expressions are processed more rapidly and accurately than negative or neutral ones and elicit more pleasant reactions from others (Kauschke et al., 2019; Mancini et al., 2018), which in turn has direct implications for social interactions. From a neurobiological standpoint, the fact that sleep deprivation has been shown to dampen facial expressions of emotion but amplify activation in emotional brain regions (Yoo et al., 2007) suggests a potential “de-coupling” effect of sleep loss where internal experiences of emotion may become disconnected from outward emotional behaviors. For example, Schwarz et al. (2013) showed that voluntary muscular facial reactions in response to both emotion-compatible and incompatible stimuli are slowed after a night of restricted sleep, whereas subjective ratings of emotion were generally unaffected. Thus, sleep loss may produce a general dampening effect on behavioral responses rather than disinhibition of top-down emotional control.

Despite the novelty of these data, our sample size was small (partly due to some limitations of the FaceReader software) which may have precluded our ability to detect other meaningful relationships. Our assessment of social problems was also limited to parental perceptions during the past 6 months which might not align with child reports. Although we utilized two carefully selected sets of counterbalanced emotional stimuli, the sleep-restricted assessment followed the rested assessment which may have introduced practice effects. Lastly, a range of additional factors not examined could also have impacted social problems at T2.

Overall, findings indicate the importance of studies exploring how sleep affects multiple facets of children’s socioemotional behavior and provide insight into one potential mechanism through which inadequate sleep might undermine social functioning. Positive emotions in particular have received limited empirical attention in relation to children’s sleep and peer relationships despite established connections with both domains of functioning. Future studies that examine these relationships within a developmental framework are needed.


  1. Achenbach, T., & Rescorla, L. (2001). Manual for the ASEBA school-age forms & profiles: Child behavior checklist for ages 6–18, teacher’s report form, youth self-report. In An integrated system of multi-informant assessment.

  2. Alfano, C. A., Bower, J. L., Harvey, A. G., Beidel, D. C., Sharp, C., & Palmer, C. A. (2020). Sleep restriction alters children’s positive emotional responses, but effects are moderated by anxiety. Journal of Child Psychology and Psychiatry and Allied Disciplines, 61(10).

  3. Baum, K. T., Desai, A., Field, J., Miller, L. E., Rausch, J., & Beebe, D. W. (2014). Sleep restriction worsens mood and emotion regulation in adolescents. Journal of Child Psychology and Psychiatry and Allied Disciplines, 55(2).

  4. Brand, S., Hatzinger, M., Stadler, C., Bolten, M., von Wyl, A., Perren, S., von Klitzing, K., Stadelmann, S., & Holsboer-Trachsler, E. (2015). Does objectively assessed sleep at five years predict sleep and psychological functioning at 14 years? - Hmm, yes and no! Journal of Psychiatric Research, 60.

  5. Cairns, R. B., Leung, M. C., Buchanan, L., & Cairns, B. D. (1995). Friendships and social networks in childhood and adolescence: Fluidity, reliability, and interrelations. Child Development, 66, 1330–1345. PMID: 7555219.

    Article  Google Scholar 

  6. Cootes, T.F., Edwards, G. & Taylor, C.J. (1998). Active appearance models. In H.Burkhardt and B. Neumann (eds.), 5th European Conference on Computer Vision, 2, 484–498. Springer, Berlin, 1998.

  7. Dekovic, M., & Gerris, J. R. M. (1994). Developmental analysis of social, cognitive and behavioral differences between popular and rejected children. Journal of Applied Developmental Psychology, 15, 367–386.

    Article  Google Scholar 

  8. Ekman, P., Roper, G., & Hager, J. C. (1980). Deliberate facial movement. Child Development, 51(3).

  9. Foley, J. E., & Weinraub, M. (2017). Sleep, affect, and social competence from preschool to preadolescence: Distinct pathways to emotional and social adjustment for boys and for girls. Frontiers in Psychology, 8, Article 711.

  10. Gosselin, P., Maassarani, R., Younger, A., & Perron, M. (2011). Children’s deliberate control of facial action units involved in sad and happy expressions. Journal of Nonverbal Behavior, 35(3).

  11. Hubbard, J. A. (2001). Emotion expression processes in children’s peer interaction: The role of peer rejection, aggression, and gender. Child Development, 72(5).

  12. Hsee, C. K., Hatfield, E., Carlson, J. G., & Chemtob, C. (1993). Emotional contagion and its relationship to mood. In E. Hatfield, J. T. Cacioppo, & R. L. Rapson (Eds.), Emotional contagion (pp. 150–152). Cambridge University Press.

    Google Scholar 

  13. Kauschke, C., Bahn, D., Vesker, M., & Schwarzer, G. (2019). The role of emotional valence for the processing of facial and verbal stimuli—positivity or negativity bias? Frontiers in Psychology, 10, 1654.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2005). International affective picture system (IAPS): Affective ratings of pictures and instruction manual. In Tech. Report A-6.

  15. Larose-Grégoire, É., Bernier, A., Dirks, M., & Perrier, R. (2018). Associations between sleep consolidation in infancy and peer relationships in middle childhood. Social Development, 27(2).

  16. Mancini, G., Biolcati, R., Agnoli, S., Andrei, F., & Trombini, E. (2018). Recognition of facial emotional expressions among italian pre-adolescents, and their affective reactions. Frontiers in Psychology, 9, 1303.

    Article  PubMed  PubMed Central  Google Scholar 

  17. McGlinchey, E. L., Talbot, L. S., Chang, K., Kaplan, K. A., Dahl, R. E., & Harvey, A. G. (2011). The effect of sleep deprivation on vocal expression of emotion in adolescents and adults. Sleep, 34, 1233–1241.

    Article  PubMed  PubMed Central  Google Scholar 

  18. McMakin, D. L., Dahl, R. E., Buysse, D. J., Cousins, J. C., Forbes, E. E., Silk, J. S., Siegle, G. J., & Franzen, P. L. (2016). The impact of experimental sleep restriction on affective functioning in social and nonsocial contexts among adolescents. Journal of Child Psychology and Psychiatry and Allied Disciplines, 57(9).

  19. Minkel, J., Htaik, O., Banks, S., & Dinges, D. (2011). Emotional expressiveness in sleep-deprived healthy adults. Behavioral Sleep Medicine, 9(1).

  20. Reddy, R., Palmer, C. A., Jackson, C., Farris, S. G., & Alfano, C. A. (2017). Impact of sleep restriction versus idealized sleep on emotional experience, reactivity and regulation in healthy adolescents. Journal of Sleep Research, 26(4).

  21. Schwarz, J. F., Popp, R., Haas, J., Zulley, J., Geisler, P., Alpers, G. W., Osterheider, M., & Eisenbarth, H. (2013). Shortened night sleep impairs facial responsiveness to emotional stimuli. Biological Psychology, 93, 41–44.

    Article  PubMed  Google Scholar 

  22. Vaughn, B. E., Elmore-Staton, L., Shin, N., & El-Sheikh, M. (2015). Sleep as a support for social competence, peer relations, and cognitive functioning in preschool children. Behavioral Sleep Medicine, 13(2).

  23. Viola, P. & Jones, M. (2001). Rapid object detection using a boosted cascade of simple features. Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 511–518.

  24. Yoo, S. S., Gujar, N., Hu, P., Jolesz, F. A., & Walker, M. P. (2007). The human emotional brain without sleep–a prefrontal amygdala disconnect. Current Biology, 17, R877–R878.

    Article  PubMed  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Candice A. Alfano.

Ethics declarations


This research was supported by a grant (#R21MH099351) from the National Institute of Mental Health (NIMH) awarded to the first author and a grant (#1813574) from the National Science Foundation awarded the last and first authors.

Data Availability

The data for this work can be found here:

Code Availability

SPSS version 27.0

Ethics Approval

All procedures were performed in accordance with the ethical standards of the University of Houston and the 1964 Helsinki declaration and its later amendments and ethical standards. The study was approved by the Institutional Review Board at the University of Houston.

Conflicts of Interest

The authors declare no competing interests.

Informed Consent

was obtained from all individual participants included in the study.

Additional information

Handling Editor: Aric Prather

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Alfano, C.A., Kim, J., Cifre, A.B. et al. Children’s Emotional Expressivity After Sleep Restriction Forecasts Social Problems Years Later. Affec Sci (2021).

Download citation


  • Sleep
  • Emotion
  • Emotional expression
  • Social functioning