Maternal Education and Sleep Quality Moderate the Associations Between Infant Sleep and Cognition

• In Event: 2-089 - Sleep and Cognition from Infancy to Middle Childhood: The Role of Socioeconomic Factors

Integrative Statement

Introduction: Links between sleep and cognition have been found in experimental studies of children (e.g., Sadeh et al., 2003). Fewer studies have investigated these links in infancy. Additionally, few studies have examined whether infant sleep interacts with aspects of the caregiving environment to predict cognition (e.g., Bernier et al., 2014). The current study tests whether infant sleep quality interacts with various maternal indicators of caregiving quality (i.e., years of education, depressive symptoms, sleep quality) to predict infant cognitive outcomes.
Methods: Data come from a longitudinal study of sleep development in African-American infants. 95 parent-child dyads (54% male) were visited in their homes at two timepoints. At 3 months, infant sleep was objectively measured using videosomnography and actigraphy, and mothers completed questionnaires regarding demographics, depressive symptoms (CES-D; Radloff, 1977), and their own sleep quality (PSQI; Buysse et al., 1988). At 6 months, infants completed cognitive assessments, including measures of general cognitive ability (Bayley, 2006) and sustained attention (Perry et al., 2016).
Infant sleep quality was modeled as two correlated latent variables representing sleep consolidation (longest sleep period, nighttime sleep ratio, number of daytime naps) and regulation (number of nighttime parental interventions, self-soothed wakes, total wake time). The measurement model (Figure 1) fit the data well, χ2 (8) = 11.57, p = .17, CFI = .97, RMSEA = .07, SRMR = .08. Subsequently, three structural equation models were estimated predicting Bayley scaled scores and sustained attention from: a) latent sleep consolidation and regulation, b) one of three maternal variables, and c) two latent interaction terms (consolidation X maternal characteristic; regulation X maternal characteristic).
Results: Our findings revealed that infant sleep quality interacted with maternal education and sleep quality to predict cognition. There were no significant direct or interactive effects for maternal depression. Regarding maternal education (Figure 2a), the relationship between infant sleep regulation and sustained attention was significant and positive for infants whose mothers had a college education (b = 4.05, p < .001). For infants whose mothers had only a high school education, there was no significant relationship between sleep regulation and sustained attention (b = -1.59, p = .22). Regions of significance analyses indicated that the relationship between sleep regulation and looking time became significant when mothers had obtained 14.4 years of education (-0.14 SD), which is approximately equivalent to two years of college. Regarding maternal sleep quality (Figure 2b), we found that the relationship between infant sleep consolidation and Bayley scaled scores was significant and positive, but only for infants whose mothers reported poor (-1SD) sleep quality (b = 1.44, p = .04). For infants whose mothers had better sleep quality (+1 SD), the relationship between sleep and Bayley scores was not significant (b = -.77, p = .39).
Together, these findings suggest that higher quality infant sleep may both augment impacts of positive environments (e.g., high maternal education), as well as buffer against impacts of negative environments (e.g., poor maternal sleep quality) on infants’ cognitive functioning.

Authors

• Marie Camerota, UNC-Chapel Hill
  Presenting Author

• Noa Gueron-Sela, Ben-Gurion University of the Negev
  Non-Presenting Author

• Cathi Propper, University of North Carolina at Chapel Hill
  Non-Presenting Author