The Relationship Between Executive Functions and Language Abilities in Children: A Latent Variables Approach

Abstract

Purpose: We aimed to outline the latent variables approach for measuring nonverbal executive function (EF) skills in school-age children, and to examine the relationship between nonverbal EF skills and language performance in this age group. Method: Seventy-one typically developing children, ages 8 through 11, participated in the study. Three EF components, inhibition, updating, and task-shifting, were each indexed using 2 nonverbal tasks. A latent variables approach was used to extract latent scores that represented each EF construct. Children were also administered common standardized language measures. Multiple regression analyses were conducted to examine the relationship between EF and language skills.Results: Nonverbal updating was associated with the Receptive Language Index on the Clinical Evaluation of Language Fundamentals-Fourth Edition (CELF-4). When composites denoting lexical-semantic and syntactic abilities were derived, nonverbal inhibition (but not shifting or updating) was found to predict children's syntactic abilities. These relationships held when the effects of age, IQ, and socioeconomic status were controlled. Conclusions: The study makes a methodological contribution by explicating a method by which researchers can use the latent variables approach when measuring EF performance in school-age children. The study makes a theoretical and a clinical contribution by suggesting that language performance may be related to domain-general EFs. E xecutive functions (EFs) are a set of top-down cognitive control processes used to manage thought and behavior (Diamond, 2013;Miyake et al., 2000). EFs are crucial to the ability to adapt efficiently to changes in the environment (Huizinga, Dolan, & van der Molen, 2006;Zelazo, Müller, Frye, & Marcovitch, 2003), and to the ability to manage basic daily tasks, such as planning, decision making, and problem solving (Friedman et al., 2006;Miyake et al., 2000). It is not surprising then that there is a wealth of literature linking EFs and broad quality-of-life outcomes, including academic achievement (e.g., Best, Miller, & Naglieri, 2011;Blair & Razza, 2007) and social-emotional development (Broidy et al., 2003;Ferrier, Bassett, & Denham, 2014) There is a great deal of interest in accurately measuring EF skills in our field, both for the purposes of documenting EF deficits in impaired populations and for the purposes of possibly influencing language outcomes through targeting EF skills. In the present study, we examined EF performance in a sample of monolingual typically developing children, with the following three goals. The first goal was a methodological one: We aimed to document the use of a latent variables approach when measuring nonverbal EF performance in children. The second goal was a clinical one: We aimed to examine whether children's performance on common standardized language measures is associated with EFs. The third goal was a theoretical one: We aimed to examine whether domain-general EF skills (as indexed by performance on nonverbal EF tasks)...

Domain‐General Cognitive Ability Predicts Bilingual Children's Receptive Vocabulary in the Majority Language

“…To assess vocabulary, a definitional vocabulary test was used in which the participants had to mention the function or relevant context for an item presented to them. The use of one response inhibition test may have limited any findings because children's behavior on executive function tasks is variable, and single tests are seldom pure measures of assumed underlying constructs (Kaushanskaya et al., ; Miyake et al., ). In addition, a definitional vocabulary test might be relatively sensitive to domain‐specific knowledge and less to domain‐general cognitive ability, unlike the simple matching tasks used in our study to measure receptive vocabulary and by Bohlman et al.…”

“…First, it allowed us to reduce the number of correlational analyses from 54 (three testing sessions, three tests, both directions) to 18. Second, as explained earlier, working memory and attention are closely related (Awh et al, 2006;Cowan, 1995;Cowan et al, 2005;Engle et al, 1999) and principal component analysis may tap into a shared underlying construct, which is better represented by an index based on multiple tasks than separate tasks (Kaushanskaya et al, 2017;Miyake et al, 2000). As the sample size was too small for structural equation modeling and the extraction of a latent variable, principal component analysis was used.…”

“…() found that a latent executive function variable at the beginning of preschool predicted children's receptive vocabulary at the end of preschool. A latent executive function variable is based on multiple executive function measures and does justice to the multicomponent structure of executive function while taking into account task variability and impurity (Kaushanskaya et al., ; Miyake et al., ). In Weiland et al.’s study, receptive vocabulary did not predict executive function, in contrast to other longitudinal research (Fuhs & Day, ; Kuhn et al., ).…”

“…Various studies have indeed found that children's performance on executive function tasks is related to their language outcomes (Kapa & Colombo, 2014;Kaushanskaya, Park, Gangopadhyay, Davidson, & Weismer, 2017;Lonigan et al, 2016;Montgomery & Evans, 2009;Verhagen & Leseman, 2016;Yoshida et al, 2011), but only a handful of studies have investigated relationships between cognitive ability and vocabulary development and used a longitudinal design. The strength of a longitudinal approach is that sequential relationships can be identified, and temporal precedence can be established.…”

Variability of Executive Function Performance in Preschoolers with Developmental Language Disorder

“…Specifically, working memory was associated with children's receptive language abilities and inhibition skills were related to lexical-semantic and syntactic abilities. 12 The relations observed between executive function and language abilities are especially interesting because the executive function measures were nonverbal, meaning that the connection to language was not due to overt linguistic demands on executive function measures.…”