Examining How Pattern, Spatial, Executive Function, and Language Skills Contribute to Preschoolers' Math Knowledge

• In Event: More Than Just Numbers: Varied Predictors of Mathematics Knowledge

Abstract

Math knowledge develops rapidly in early childhood, varies between children (Starkey, Klein, & Wakeley, 2004), and is strongly predictive of both later math and reading achievement (Duncan et al., 2007). While theory and research on math has concentrated primarily on contributions of children’s numerical knowledge (Sarama & Clements, 2004), emerging evidence shows that early pattern, spatial, and general cognitive skills (e.g., language and EF) predict concurrent and later math skills (e.g., Bull et al., 2008; Duncan et al., 2007; Mix & Cheng, 2012; Rittle-Johnson, Fyfe, Hofer, & Farran, 2015). Little is known, however, about the association between these skills, nor how together they explain variation in math knowledge.
Hypothesis 1: Given the similarities between pattern, spatial, language, and EF skills in preschoolers, these skills should be moderately to strongly correlated.
Hypothesis 2: Pattern and spatial skills in the beginning of the preschool year were expected to significantly and independently predict math knowledge at the beginning and end of the year after controlling for other cognitive skills.
Seventy-three 4- to 5-year-old children (Mean age = 4 years, 7 months) were recruited from 6 preschools, and assessed at the beginning and end of the school year. The Research-Based Early Mathematics Assessment (REMA)-Short Form assessed math skill (Weiland et al., 2012), and was chosen because it measures numeracy and shape knowledge. Spatial skills examined included form perception (assessed via the Position in Space subtest from the Developmental Test of Visual Perception; Hammill, Pearson & Voress, 1993), spatial visualization (using Block Design, WPPSI-IV; Wechsler, 2012) and visual-spatial working memory, measured via the Corsi Block Tapping Task (Pagulayan et al., 2006). Children’s pattern skills were assessed using a validated measure of repeating pattern knowledge (Rittle-Johnson et al., 2013), and a novel patterning assessment created using online resources for classroom teachers. The pattern and spatial measures were standardized and averaged to create a pattern and a spatial composite score. General cognitive ability measures included children’s verbal ability (NIH Toolbox Picture Vocabulary Test) and their verbal working memory (WM; measured via the backward digit span task from the WISC-III (Wechsler, 1991).
In support of Hypothesis 1, the pattern and spatial composites, verbal ability and WM were significantly correlated with each other (see Table 1). In support of Hypothesis 2, pattern and spatial composites each predicted mathematics knowledge at the beginning of preschool and end of preschool (see Table 2), together explaining an additional 13% of variance in math knowledge per model. Verbal working memory, was only a significant predictor of math at the beginning of the school year. However, after controlling for beginning math knowledge, only the spatial composite measure was a unique predictor of end of preschool math knowledge, and this relation was weaker.
Thus, while moderately correlated, preschoolers’ pattern, spatial, and working memory skills each uniquely predicted children’s concurrent and later math knowledge. Theories and standards for early math development should be expanded to incorporate a role for patterning, spatial, and executive function skills.

Authors

• Erica Leigh Zippert, Vanderbilt University

• Bethany Rittle-Johnson, Vanderbilt University