Intra-Individual Variability in Sleep Predicts Cognitive Performance in School-Aged Children: A Longitudinal Study

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Cognitive Processes

Abstract

Sleep has an extensive influence on child development (Spruyt, 2019) including cognitive functioning. However, most studies to date have been cross-sectional in nature (Astill, Van der Heijden, Van IJzendoorn, & Van Someren, 2012; Short et al., 2018). Consequently, the possible long-term effects of sleep are largely unknown. Additionally, few studies have examined whether night-to-night variability in sleep duration (VSD) may bear on child cognitive functioning. Such an impact seems likely, given that heightened variability in sleep is associated with lower white matter integrity in adolescents (Telzer et al., 2015) and lower verbal abilities in school-aged children (Buckhalt, El-Sheikh, & Keller, 2007; Buckhalt, El-Sheikh, Keller, & Kelly 2009). This study examined the links between VSD in school-aged children and their cognitive performance one year later. We anticipate that after a one-year span, greater night-to-night VSD will be associated with lower cognitive performance.

Ninety-nine typically developing children (55 females) were studied in two consecutive years. In the fourth grade (M = 9.94 years, SD = 0.29), children’s reading (WIAT-II word reading), planning (NEPSY-II Tower of LondonDX), inhibition (D KEFS Stroop), working memory (Digit Span), and cognitive flexibility skills (Dimensional Change Card Sort) were assessed. Their sleep was also objectively measured using a Mini-Mitter® Actiwatch worn on their non dominant wrist for 3 to 7 nights (M = 6.03) and analyzed using a threshold of 80 activity counts per epoch. VSD was calculated for each participant by computing the standard deviation of their night-to-night sleep duration. One year later, the same cognitive measures were re-administered when participants were in the fifth grade.

Only reading and planning skills were significantly correlated with VSD (p < .05). These two cognitive indicators were then submitted to hierarchical regression analyses, which revealed that after controlling for participants’ family socioeconomic status (SES) and sex, fourth-grade VSD was a significant predictor of fifth grade reading skills, β = −0.25, p = 0.030 as well as planning skills, β = −0.33, p = 0.004. In a second set of regressions, corresponding fourth-grade cognition scores were also covaried. Fourth-grade VSD was not a reliable predictor of reading skills in fifth grade after accounting for previous reading skills, β = −0.19, p = 0.081. However, fourth-grade VSD was a significant predictor of fifth-grade planning skills above and beyond participants’ SES, sex, and previous planning skills, β = −0.24, p = .028. Therefore, greater night-to-night sleep variability in children was predictive of lower reading skills and planning skills one year later, and also predicted less improvement in planning skills across a period of one year.

This study underscores the presence of some long-term associations between sleep and cognition in school-aged children. It also highlights the importance of studying diverse aspects of sleep, as different indicators of sleep may bear differently on the cognitive functioning of children. Further research is necessary to identify the social, cognitive and neural factors that may account for the associations between VSD and child cognition.

Authors

• Anna-Francesca Boatswain-Jacques, McGill University
  Presenting Author

• Charlotte Dusablon, University of Montreal
  Non-Presenting Author

• Annie Bernier, University of Montreal
  Non-Presenting Author