The contribution of executive function and social understanding to preschoolers’ letter and math skills

Highlights

• We examined underlying processes of preschoolers’ academic school readiness.

• A battery of executive function and social understanding tasks were administered.

• Four structural equation models were examined and compared.

• Working memory uniquely predicted preschoolers’ early letter and math skills.

Abstract

The influence of executive function and social understanding on letter and math skills was examined in 129 3–5-year-olds. Tasks were administered to measure working memory, inhibition, social understanding, letter and math skills, and vocabulary. Using latent variable analyses, multiple models were compared in order to examine the influence of executive function and social understanding on participants’ emerging academic skills. In the best-fitting model, working memory contributed to letter and math skills, over and above inhibition, social understanding, age, and vocabulary. Inhibition and social understanding did not uniquely contribute to letter and math skills, but significant relations were found among working memory, inhibition, and social understanding. Findings are discussed with respect to improving ways to examine the complex relations among preschoolers’ executive function, social understanding, and school readiness skills.

Introduction

Preschoolers’ rudimentary academic skills, such as knowledge of letters and numbers, may be important for later school achievement because they create a strong motivational basis for future learning and facilitate the acquisition of further academic skills (Duncan et al., 2007, Pagani et al., 2010, Romano et al., 2010). Empirical evidence for this claim comes from a growing body of research showing that early academic skills predict later school achievement (Lemelin et al., 2007; for a meta-analysis, see La Paro & Pianta, 2000). The majority of these studies have focused on young children's early literacy and math skills, often assessed in terms of identifying letters, words, numbers, counting, and shapes. For example, Duncan et al. (2007) analyzed six longitudinal data sets and found that preschoolers’ math skills, such as their knowledge of numbers, were the most powerful predictors of later learning (average β = .34), followed by preliteracy skills, such as letter knowledge (average β = .17).

In addition to specific academic skills, recent research on school readiness and achievement has highlighted the importance of more domain-general cognitive skills (Blair, 2002, Bodrova and Leong, 2006). Executive function (EF) and, to a lesser extent, social understanding (SU) have received attention (Astington and Pelletier, 2005, Blair and Razza, 2007, Meltzer, 2007, Monette et al., 2011, Müller et al., 2008, Welsh et al., 2010). EF refers to an interrelated set of higher cognitive processes used in the control of action and thought (Garon et al., 2008, Zelazo and Müller, 2010), whereas SU (also referred to as social cognition or theory of mind) refers to the attribution of mental states to self and others that can be used to explain and to predict behavior (Astington, 1993, Carpendale and Lewis, 2006). Research suggests that a functional relation exists between EF and SU because (a) both undergo dramatic developmental changes during the preschool years, (b) both have neural underpinnings in the frontal lobes, and (c) impairments in both are implicated in various developmental disorders (Sabbagh et al., 2008, Zelazo and Müller, 2010). Moreover, significant correlations between EF and SU tasks have been found both in typically (Moses & Tahiroglu, 2010) and in atypically (Pelicano, 2010) developing children, even after controlling for age, verbal ability, and IQ.

Whereas the relation between EF and SU in young children is well supported, the influence of each on children's school readiness is inconsistent across studies, especially in terms of their relative contribution to academic performance. Therefore, the goal of the present study was to examine the combined and unique influence of EF and SU on preschoolers’ school readiness skills, specifically letter and math skills. We also sought to better explain variance in preschoolers’ academic skills by using a latent variable approach.

Factor analytic studies suggest that EF in school-age children includes component processes of inhibition, working memory/updating, and shifting/flexibility (Garon et al., 2008, Huizinga et al., 2006, Lehto et al., 2003). In these studies, inhibition refers to the ability to suppress prepotent responses, working memory to the ability to monitor and revise information, and shifting to the ability to switch between multiple tasks (Miyake et al., 2000). In preschoolers, these component processes may remain relatively undifferentiated (Zelazo & Müller, 2010). Recent latent variable studies involving a variety of EF tasks have tended to support a unitary EF factor structure in preschoolers (Hughes et al., 2010, Wiebe et al., 2008, Wiebe et al., 2011, Willoughby et al., 2010, Willoughby et al., 2012a, Willoughby et al., 2012b). However, in most of these studies, a two-factor EF structure consisting of working memory and inhibition still fit the data well, but was rejected in favor of a unitary structure on grounds of parsimony. Moreover, other latent variable studies have found that a two-component EF structure with working memory and inhibition as latent factors fit the data better than a unitary EF structure both in typically (Miller, Giesbrecht, Müller, McInerney, & Kerns, 2012) and atypically developing preschoolers (Schoemaker et al., 2012). Therefore, while the structure of EF remains open to investigation, the importance of working memory and inhibition processes is well recognized in preschoolers’ EF development (Garon et al., 2008).

Composite measures of EF and different components of EF both have been linked to a variety of indicators of academic achievement in school-aged children (Best et al., 2011, Visu-Petra et al., 2011, Waber et al., 2006; for a review, see Müller et al., 2008). Less research has been conducted with preschoolers and kindergartners, but there is increasing evidence that emerging academic skills are significantly correlated with composite measures and individual components of EF in younger children (Alloway et al., 2005, Bull et al., 2011, Espy et al., 2004, Kroesbergen et al., 2009, Lan et al., 2011). Furthermore, longitudinal studies suggest that EF facilitates the acquisition of emerging academic skills (Blair and Razza, 2007, Bull et al., 2008, Clark et al., 2010, McClelland et al., 2007, Monette et al., 2011, NICHD, 2003, Welsh et al., 2010). For example, Welsh et al. (2010) found that growth in EF (i.e., working memory and attentional control) over the course of the prekindergarten year predicted (a) growth in literacy (identifying and saying letters and words) and math skills (numbers, quantities, counting, and simple arithmetic) between the beginning and end of the prekindergarten year and (b) kindergarten reading and math achievement after controlling for growth in literacy skills, math skills, and verbal ability.

Although research suggests that EF contributes to school readiness skills, findings concerning the relative contribution of different components of EF are inconsistent. There is evidence that individual differences in inhibition explain variance in preschoolers’ emerging literacy and math skills concurrently as well as prospectively (Blair and Razza, 2007, Espy et al., 2004, McClelland et al., 2007, NICHD, 2003). For example, Blair and Razza (2007) measured inhibition and shifting in preschoolers and again a year later in kindergarten, when their letter and math skills were also assessed (numbers, quantities, sizes, shapes, and basic arithmetic and graphic relations). After controlling for shifting, IQ, and verbal ability, inhibition uniquely predicted (a) concurrent letter knowledge in kindergarten and (b) both concurrent and prospective math knowledge in kindergarten. In addition, Espy et al. (2004) examined the relative contribution of inhibition, working memory, and shifting to preschoolers’ math skills (subitizing, counting, and simple arithmetic). Both inhibition and working memory accounted for unique variance in math skills after controlling for age, language skills, and maternal education. However, after controlling for the other EF components as well, only inhibition accounted for a unique portion of variance in math skills, suggesting that inhibition is particularly important for early math skills.

Other researchers, however, have emphasized the importance of working memory for school readiness and achievement (Alloway et al., 2005, Gathercole and Alloway, 2008, Gathercole et al., 2004, Passolunghi et al., 2007, Van der Ven et al., 2012). In fact, there is some evidence that in older children, working memory is more important for school achievement than inhibition. Working memory has been found to explain a larger amount of variance in school achievement than inhibition (St. Clair-Thompson & Gathercole, 2006) and to be the only component of EF that uniquely predicted mathematical ability in school-age children over and above reading ability and other components of EF (Bull and Scerif, 2001, Swanson, 2006). In a similar vein, recent studies suggest that working memory, but not inhibition, in kindergarteners makes a unique contribution to academic skills (Bull et al., 2008, Lan et al., 2011, Monette et al., 2011). For example, Monette et al. (2011) found that kindergarteners’ working memory, but not their inhibition or shifting, predicted their math achievement at the end of grade 1, even after controlling for earlier school readiness skills (colors, letters, numbers), age, maternal education, and family income. However, none of the EF components directly predicted reading and writing achievement at the end of grade 1.

Few studies have examined the relative contribution of shifting to preschoolers’ school readiness, and most studies that have included measures of shifting have not shown significant relations to school readiness (Espy et al., 2004, Monette et al., 2011). One exception is a recent study by Vitiello, Greenfield, Munis, and George (2011) showing that teachers’ ratings of preschoolers’ attention and persistence mediated the relation between shifting and school readiness (colors, letters, numbers, sizes, object comparisons, and shapes) in preschoolers in Head Start classrooms. However, the zero-order correlation between shifting and school readiness was relatively small (r = .19, p < .01). Thus, shifting may have some influence on preschoolers’ school readiness, but the influence seems to be small or indirect. Shifting may be a relatively indistinct component of EF among preschoolers, as suggested by recent latent variable studies of EF that do not identify a unique shifting component in preschoolers (Miller et al., 2012, Willoughby et al., 2010, Willoughby et al., 2012a) or elementary school children (Lee et al., 2012, Van der Ven et al., 2012). Overall, given existing evidence is inconsistent, the relative contribution of different EF components to early literacy and math skills as well as to school achievement merits further clarification.

A milestone in the development of SU emerging around 4 years of age is false belief understanding, or comprehending that less-informed people may hold and act on beliefs that are not true (Astington, 1993, Carpendale and Lewis, 2006, Wellman et al., 2001). In the classical task used to assess false belief understanding (Wimmer & Perner, 1983), children have to predict where a protagonist (Maxi) will search for an object that unbeknownst to him has been moved to a new location. Although children know where the object has been moved to, they need to realize that the protagonist will act on his false belief and search for the object in the original location. False belief understanding has been linked to language development (Milligan, Astington, & Dack, 2007), and children's language skills are important in forming interpersonal relationships that contribute to early academic success (Mashburn & Pianta, 2006).

Astington and Pelletier (2005) have suggested that SU, and false belief understanding in particular, is important for school achievement because it is associated with social maturity, collaborative learning ability, cognitive monitoring, narrative comprehension, and scientific thinking. They found that false belief understanding was related to reading ability and understanding of narrative; the latter, but not the former, relation remained significant after controlling for language competence. Similarly, Blair and Razza (2007) found that false belief understanding predicted 6-year-olds’ letter knowledge after controlling for performance on EF tasks, IQ, and verbal ability (at age 5). Furthermore, Blair and Razza (2007) found a positive zero-order correlation between preschoolers’ false belief understanding and math skills (numbers, sizes, shapes) in kindergarten, but the relation became insignificant after controlling for EF and effortful control. However, given the sparse evidence, it remains to be determined whether a functional relation exists between SU and emerging academic skills. Specifically, it is unclear whether SU accounts for unique variance in children's early letter and math skills because either no (Astington & Pelletier, 2005) or few (Blair & Razza, 2007) EF components have been accounted for in these studies.

The objective of the present study is to clarify the extent to which EF and SU contribute to letter and math skills in preschool children. As summarized, previous research has shown that both EF and SU are related to early letter and math skills. However, with one exception (Blair & Razza, 2007), previous research has not included as predictors both EF and SU. At the same time, several studies have shown that EF and SU (especially false belief understanding) are significantly correlated in preschool children (for a review, see Moses & Tahiroglu, 2010). As a consequence, the unique contribution of EF and SU to letter and math skills in preschoolers remains unclear. A further problem is that latent variable model comparisons are rarely used to assess these relations. Latent variable approaches have the advantages of (a) examining theory-driven, a priori relations between factors and (b) accounting for measurement error by extracting only the common variance shared by different tasks that are stipulated to measure the same latent factor (Bryant and Yarnold, 1994, Klem, 2000). With few exceptions (e.g., Bull et al., 2011, Lee et al., 2012, van der Sluis et al., 2007, Van der Ven et al., 2012, Willoughby et al., 2010, Willoughby et al., 2012a), latent variable model comparisons have not been used to evaluate the contributions of EF or SU to school readiness and academic skills. Accordingly, we wished to use this method to ascertain the relative contribution of different aspects of EF and SU to early letter and math skills. More broadly, the study should illuminate the relation between more general cognitive skills, such as EF and SU, and domain-specific academic skills.