Poster #10 - Testing Observations of Physical Activity and Play as Predictors of Executive Function in Early Childhood.

• In Event: 3-003 - Poster Session 09
  In Poster Session: PS 09 Section - Attention, Learning, Memory

Integrative Statement

Moderate-vigorous physical activity (MVPA) has been associated with improved cognitive development, including executive function (EF) skills, and their associated neural substrates in school-aged children (Chaddock, Pontifex, Hillman & Kramer, 2011; Donnelly et al, 2016; Hillman, Khan & Kao, 2015). Less is known about the association of MVPA and EF skills in early childhood (Carson et al., 2015; Willoughby, Wylie & Catellier, 2017). Notably, efforts to increase MVPA in school-aged children involve discrete exercise activities. In early childhood, discrete exercise may not be developmentally appropriate, as most physical activity occurs in the context of play.

Current observational tools that seek to characterize the nature of children’s play do not adequately measure the child’s physical activity intensity. To address this shortcoming, we integrated two observational tools, the Play Observation Scale (POS; Rubin, 2001) and the OSRAC-P (Brown et al., 2006) to create a hybrid observation system (i.e., the Kids Activity & Learning Study Observational Scheme; KALOS). The KALOS combined specific components of the POS (e.g., codes for play type and social composition) and OSRAC-P (e.g., codes for physical activity intensity and location). Over ten-minute observation sessions, the focal child is observed for 5 seconds, followed by 10 seconds of coding; during this time, the observer records the child’s physical activity intensity and distinguishes whether the child is participating in “play” behavior, and if so, chooses the type of play and social composition of that play among other codes.

We report data on the first cohort of participants from the Kids Activity and Learning Study. Preschool children (N=97) were observed for four sessions during free-play and completed an EF assessment within the same week. Observational data was summarized into variables describing the proportion of time spent in codes within higher-level categories. Bivariate correlations were used to examine relationships between KALOS codes and age, EF, and simple reaction time (SRT). Ordinary least squares regression models were used to test the contributions of KALOS codes (principally MVPA, social composition categories of play, play types, and interactions between MVPA x social composition and MVPA x play types) on EF while controlling for relevant covariates.

Bivariate associations among KALOS codes, EF skills, and covariates are summarized in Table 1. KALOS codes were related in expected ways. The only KALOS code that was related to EF was group play (r=0.20, p = 0.02). The unique effect of group play on EF skills persisted even after adjustment for covariates (β = 0.32, p = 0.01; see Table 2). There was no evidence that observed MVPA exerted unique effects for EF skills, nor did it moderate the effects of any social composition or play type codes. These results are based on the first cohort of children and will be updated to include the second cohort of children (minimum N = 160) prior to the SRCD conference. The KALOS may be appropriate for researchers who seek to test the unique and potentially combined contributions of play contexts and physical activity intensities in early childhood.

Authors

• Amanda M. Jones, RTI International
  Presenting Author

• Amanda Catherine Wylie, Research Triangle Institute International
  Non-Presenting Author

• Michael T Willoughby, RTI International
  Non-Presenting Author