Working memory and social functioning in children

https://doi.org/10.1016/j.jecp.2013.03.002Get rights and content

Abstract

This study extends previous research and examines whether working memory (WM) is associated with multiple measures of concurrent social functioning (peer rejection, overall social competence, relational aggression, physical aggression, and conflict resolutions skills) in typically developing fourth- and fifth-grade children (N = 116). Poor central executive WM was associated with both broad social impairments (peer rejection and poor overall social competence) and specific social impairments (physical aggression, relational aggression, and impaired conflict resolution skills); poor verbal storage was associated only with greater peer rejection, and spatial storage was not associated with any measures of social impairment. Analyses also examined whether specific impairments in aggressive behavior and conflict resolution skills mediated the association between central executive and broad measures of social functioning. Greater physical aggression and impaired conflict resolution skills were both significant mediators; relational aggression was not. Implications for theory and future research are discussed.

Introduction

Researchers have become increasingly interested in the role of executive functioning (EF) in the social competence of children (e.g., Barkley, 1997, Eslinger et al., 2004, Jurado and Rosselli, 2007, Yeates et al., 2007). Executive functioning generally is used as an umbrella term that describes a set of interrelated cognitive abilities associated with goal- and future-directed behavior (Jurado & Rosselli, 2007). Indeed, some empirical research supports a link between EF deficits and impaired social functioning among clinical populations (Clark et al., 2002, Fahie and Symons, 2003, Happé et al., 2006, Muscara et al., 2008). Unfortunately, limitations in the current literature, including a reliance on clinical samples and the use of composite measures of both EF and social competence, have prevented a full understanding of the ways in which EF may affect social functioning. To address these limitations and extend prior research, this study examined whether working memory (WM), a specific component of EF, is associated with multiple aspects of social functioning in a sample of typically developing children. In addition, this study considered whether aggression and impairment in conflict resolution skills would explain (i.e., mediate) the relation between poor WM and broad measures of impaired social functioning.

One of the major limitations of existing research examining EF and social functioning is the focus on processes in clinical populations. However, EF deficits are observed across a range of clinical presentations and are not specific to any one clinical diagnosis (Bayliss et al., 2005, Sergeant et al., 2002, Stefanatos and Baron, 2007, Willcutt et al., 2001). In addition, it is argued that EF ability may best be understood dimensionally (Brocki and Bohlin, 2004, Roberts and Pennington, 1996), with some children from clinical populations falling at the extreme low end. Currently, it is unclear whether the implications of EF for social adjustment are consistent across the entire spectrum of impairment levels or are observed only among cognitively impaired clinical samples. Importantly, from a developmental psychopathology perspective, psychopathology is best understood as a deviation from, or an exaggeration of, typical development (Rutter & Sroufe, 2000). Thus, understanding processes within typically developing children provides a necessary framework that can inform predictions regarding clinical samples.

In addition, the majority of studies examining whether EF impairments are implicated in social deficits have used composite measures of EF based on a combined score from several individual EF tests (see Fahie and Symons, 2003, Nigg et al., 1999, for exceptions). With this approach, some evidence does support a link between impaired EF and poor social functioning (Clark et al., 2002, Fahie and Symons, 2003, Happé et al., 2006, Huang-Pollock et al., 2009, Muscara et al., 2008, Rinsky and Hinshaw, 2011); whereas other work fails to support the notion that EF deficits delineate a more socially impaired subgroup of children (Biederman et al., 2004, Diamantopoulou et al., 2007, Wåhlstedt et al., 2008).

Although several studies have used composite measures of EF, there continues to be a debate over whether EF should be viewed as a unitary construct or is better conceptualized as a set of complementary but distinct cognitive abilities (Jurado & Rosselli, 2007). These more specific cognitive abilities include WM, set shifting, inhibition, fluency, and planning. Although EF may represent a single latent construct in preschool children (Wiebe, Espy, & Charak, 2008), exploratory factor analyses of EF tests in school-age children or adults typically identify multiple component functions (e.g., Levin et al., 1996, Nigg, 2006, Willcutt et al., 2005). As a result, composite measures of EF might not provide adequate information about the specific cognitive processes that relate to social functioning. For instance, Clark and colleagues (2002) found that impaired EF, based on tests measuring planning and strategy generation, was associated with greater social impairment in a sample of children with and without disruptive behavior disorders. However, Wåhlstedt and colleagues (2008) assessed EF deficits using tests that measure WM, inhibition, and verbal fluency and failed to find that EF deficits delineated a more socially impaired subgroup of children with elevated attention deficit/hyperactivity disorder (ADHD) symptoms. Huang-Pollock and colleagues (2009), on the other hand, did find that EF impairment, measured based on tests of verbal WM, inhibition, and planning, was associated with certain negative social behaviors among children with and without ADHD. Unfortunately, the variability in the specific cognitive abilities measured when assessing EF (and the tests used to measure those abilities) makes it very difficult to identify specifically why some studies find a link between EF and social abilities and other do not. In addition to variability in the measurement of EF, these studies also differ in the specific characteristics of the sample used and in the data analytic approach, which may also affect differences in findings. In fact, in one of the only studies to consider individual measures of EF, Nigg and colleagues (1999) found that three aspects of EF (inhibitory control, visual–spatial ability, and verbal fluency) were differentially related to social competence and adjustment in elementary school children. Thus, the use of composite EF scores may mask differential associations between specific EF abilities and social functioning.

Theoretical models and hypotheses have emphasized that WM may be one of the cognitive abilities of EF that is particularly important to social functioning (Barkley, 1997, Barrett et al., 2004). WM is a limited-capacity process for short-term storage, monitoring, and manipulation of information (Baddeley, 1992). Baddeley’s (1986) model of WM suggests that it is composed of a verbal storage system, a visuo-spatial storage system, and a central executive that controls and manipulates stored information and incorporates retrieved information from long-term memory. Although multiple models of WM have been proposed (e.g., Baddeley, 1986, Engle et al., 1999), these models generally suggest that WM includes a central component responsible for processing and coordinating stored information (e.g., central executive) and separate verbal and spatial storage systems. Common verbal and spatial storage WM tasks require individuals to store in short-term memory and immediately recall temporally ordered linguistic or spatially presented information. Verbal and spatial central executive WM tasks, on the other hand, require both maintenance of presented information and manipulation such as updating or reordering (Martinussen, Hayden, Hogg-Johnson, & Tannock, 2005). Models of EF have highlighted the role of WM as important for the execution of goal-directed behavior and for both retrospective and prospective thinking (Barkley, 1997). WM ability also is related to a variety of processing outcomes, including the ability to keep goal-related representations actively in mind, engage in a deliberate search of memory for additional goal-related information, and select a behavioral response based on available information (Barrett et al., 2004). Thus, children with poor WM may have greater difficulty in keeping social goals in mind, incorporating multiple pieces of social information, and thinking through social responses. These impairments are likely to lead to difficulty in effectively navigating social situations and to more negative interactions with peers.

Despite the proposal that WM should be important to social functioning, few studies have examined this link empirically. Research does suggest that WM ability predicts greater achievement of social developmental milestones in a non-clinical sample (Alloway et al., 2005) and overall social problems (e.g., disliked, dependent, lonely, clumsy) in children with and without ADHD (Kofler et al., 2011). However, an additional limitation of research focusing on WM (as well as on EF more broadly) is that studies have used broad measures of social competence. Many researchers have emphasized that social competence is composed of multiple components (Bierman, 2004, Cavell, 1990). For instance, Cavell (1990) suggested that social competence is composed of global measures of social adjustment, such as peer rejection, as well as the specific social behaviors exhibited by children and their ability to adapt behavior in accordance with shifting social demands. Thus, overall ratings of social impairment might not fully characterize the social competence of children. In addition, reliance on broad measures fails to provide adequate insight into the reasons why impaired WM may relate to negative regard by peers or a failure to have positive social relationships. Differences in how social competence is defined also may contribute to the disparate results in studies examining composite EF and social competence (e.g., Biederman et al., 2004, Clark et al., 2002, Fahie and Symons, 2003, Muscara et al., 2008, Rinsky and Hinshaw, 2011, Wåhlstedt et al., 2008). Accordingly, in this study, we considered two broad measures of social functioning, peer rejection and overall social competence, as well as two specific aspects of social behavior that may be linked to WM impairment: conflict resolution skills and aggressive behavior.

Barkley’s (1997) model of EF suggests that WM deficits may limit the acquisition and use of positive social skills because these skills require future-oriented thinking, references to social norms, and an understanding of cause and effect. One positive social skill that is likely to rely heavily on future-oriented thinking and consideration of multiple pieces of social information is conflict resolution skills or the ability to effectively resolve conflicts with peers. To date, research has not directly tested the association between WM and conflict resolution skills. However, one study investigating the association between language competence and social cognitive skills did find that in a clinic-referred sample of 7- to 14-year-old children, WM was positively correlated with children’s ability to identify the problem in hypothetical social vignettes and to select and evaluate social responses (Zadeh, Im-Bolter, & Cohen, 2007), suggesting that WM may be implicated in effective social problem solving.

Additional evidence suggests that poor WM also may relate specifically to aggressive behavior. Rapport and colleagues’ (2001) model of WM proposes that deficits in WM may lead to subsequent impairments in inhibition that result in difficulty with behavioral regulation and low frustration tolerance (Rapport, Chung, Shore, & Isaacs, 2001). Indeed, evidence does find that persistently physically aggressive boys display greater impairments in WM relative to non-aggressive and occasionally aggressive boys (Séguin, Nagin, Assaad, & Tremblay, 2004), suggesting that WM deficits are linked to physical aggression. Importantly, recent research suggests that aggression can take on both physical and relational forms. Physical aggression harms others through physical means (e.g., hitting) and is more commonly observed in males (Card, Stucky, Sawalani, & Little, 2008). Relational aggression harms others through manipulation of relationships (e.g., intentionally excluding a peer, spreading rumors, gossiping; Crick & Grotpeter, 1995) and is more commonly used by females relative to physical aggression (Card et al., 2008). To our knowledge, research has not considered whether poor WM also relates to relational aggression. Unlike physical aggression, the use of relational aggression may require a certain level of goal-directed planning to carry out. In addition, some have argued that relational aggression may be a more skillful form of aggression and may be influenced by motivations for social status and resource control (see Heilbron & Prinstein, 2008, for a review). If this is the case, then cognitive impairments, including WM, might not be implicated in relational aggression. Indeed, Diamantopoulou and colleagues (2007) found that an aggregate measure of EF deficits predicted greater physical aggression for children high in ADHD symptoms but was not related to relational aggression among 8- and 9-year-old children. Although this study included a measure of WM in the EF composite, the independent role of WM has not been considered.

As noted above, current investigations also have failed to consider more complex models that identify mechanisms explaining the relation between aspects of EF and social functioning. Yeates and colleagues (2007) proposed a heuristic model of social competence that incorporates impairment in EF and provides a useful framework for investigating the relation between WM and social competence. Specifically, this model suggests that children with poor EF are more likely to be socially maladjusted, including being more rejected and less accepted by peers, because they have specific impairments in social cognition (e.g., problem solving, emotional reasoning) and display greater negative social behaviors (e.g., aggression, social withdrawal). When applied specifically to WM, this model suggests that impaired conflict resolution skills and higher rates of aggression may explain why impaired WM is associated with broad measures of poor social functioning. Indeed, limited evidence finds a link between poor WM and impaired problem solving and greater aggression (Diamantopoulou et al., 2007, Séguin et al., 2004, Zadeh et al., 2007), and other research finds that these specific social impairments are associated with negative peer regard and more global measures of impaired social functioning (e.g., Crick, 1996, Parker and Asher, 1993). Despite these links, research has not empirically examined whether children with poorer WM are more likely to be rejected by their peers and regarded as socially incompetent because of greater aggressive behavior and deficits in conflict resolution skills.

Thus, in an effort to address limitations in prior research, this study examined the implications of WM for social functioning within a typically developing sample of elementary school-age children. The first aim was to examine whether WM deficits were linked with impaired social functioning. Broad measures of teacher-rated peer rejection and social competence, as well as specific measures of physical and relational aggression and conflict resolution skills, were considered. Given some evidence that relational aggression may be a more skillful form of aggression (Heilbron & Prinstein, 2008), it was expected that WM impairments would be more strongly implicated in physical aggression as compared with relational aggression. The second aim of this study was to test a mediation model based on Yeates and colleagues’ (2007) heuristic model of EF and social competence. Specifically, we hypothesized that impairments in conflict resolution skills and higher rates of aggression would mediate the relation between poor WM and broad measures of social functioning.

Section snippets

Participants

Participants for this study were recruited from a larger sample of fourth- through sixth-grade students participating in a longitudinal study (N = 289 at baseline). In the larger sample, principals of local schools were approached and asked whether they would permit recruitment of participants from their schools. If principals agreed, teachers were then approached and asked whether they would allow a letter describing the study and consent form to be sent to all students in their class. All

Preliminary analyses

Preliminary analyses examined whether children participating in the current study (n = 116) and fourth and fifth graders who participated only in the larger study (n = 139) differed in social functioning. Results indicated that there were no differences in social functioning between participating and non-participating students. In addition, preliminary analyses examined whether social functioning varied as a function of demographic characteristics (age, grade, gender, parental income, ethnicity,

Discussion

The primary goal of this study was to examine the implications of WM for social functioning in typically developing children. Results supported hypotheses and found that poor WM, primarily the central executive, was associated with broad measures of peer rejection and poor overall social competence and with specific measures of conflict resolution skill deficits and physical and relational aggression. Furthermore, impairments in specific social behaviors and abilities accounted for the

Acknowledgments

This study was funded by a Child & Adolescent Psychology Training and Research Grant, awarded to the first author, and a fellowship from the University of Vermont James M. Jeffords Center, awarded to the third author. Special thanks go to the families, teachers, and school administrators who participated in this study and to Anne Brady, Mary-Louise Jacobson, Christina Moore, Elyse Ogletree, Jacqueline Shaffer, Erika Weisz and Jillian Ward for assisting in the data collection.

References (68)

  • E. Andreou

    Social preference, perceived popularity, and social intelligence: Relations to overt and relational aggression

    School Psychology International

    (2006)
  • A.D. Baddeley

    Working memory

    (1986)
  • A. Baddeley

    Working memory

    Science

    (1992)
  • R.A. Barkley

    Behavioral inhibition, sustained attention, and executive functions: Constructing a unifying theory of ADHD

    Psychological Bulletin

    (1997)
  • L. Barrett et al.

    Individual differences in working memory capacity and dual-process theories of the mind

    Psychological Bulletin

    (2004)
  • J. Biederman et al.

    Impact of executive function deficits and attention-deficit/hyperactivity disorder (ADHD) on academic outcomes in children

    Journal of Consulting and Clinical Psychology

    (2004)
  • K.L. Bierman

    Peer rejection: Developmental processes and intervention strategies

    (2004)
  • K. Björkqvist et al.

    Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression

    Aggressive Behavior

    (1992)
  • K.C. Brocki et al.

    Executive functions in children aged 6 to 13: A dimensional and developmental study

    Developmental Neuropsychology

    (2004)
  • N.A. Card et al.

    Direct and indirect aggression during childhood and adolescence: A meta-analytic review of gender differences, intercorrelations, and relations to maladjustment

    Child Development

    (2008)
  • T.A. Cavell

    Social adjustment, social performance, and social skills: A tri-component model of social competence

    Journal of Clinical Child Psychology

    (1990)
  • D. Cicchetti et al.

    Child maltreatment

    Annual Review of Clinical Psychology

    (2005)
  • C. Clark et al.

    The relationship between executive function abilities, adaptive behaviour, and academic achievement in children with externalising behaviour problems

    Journal of Child Psychology and Psychiatry

    (2002)
  • A.A. Conway et al.

    Working memory span tasks: A methodological review and user’s guide

    Psychonomic Bulletin & Review

    (2005)
  • E.L. Cowen et al.

    Long-term follow-up of early detected vulnerable children

    Journal of Consulting and Clinical Psychology

    (1973)
  • N.R. Crick

    The role of overt aggression, relational aggression, and prosocial behavior in the prediction of children’s future social adjustment

    Child Development

    (1996)
  • N.R. Crick et al.

    Relational aggression, gender, and social–psychological adjustment

    Child Development

    (1995)
  • S. Diamantopoulou et al.

    Impact of executive functioning and symptoms of attention deficit hyperactivity disorder on children’s peer relations and school performance

    Developmental Neuropsychology

    (2007)
  • T.J. Dishion

    The peer context of troublesome child and adolescent behavior

  • K.A. Dodge et al.

    Social-information-processing factors in reactive and proactive aggression in children’s peer groups

    Journal of Personality and Social Psychology

    (1987)
  • K.A. Dodge et al.

    Reactive and proactive aggression in school children and psychiatrically impaired chronically assaultive youth

    Journal of Abnormal Psychology

    (1997)
  • B.B. Efron et al.

    Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy

    Statistical Science

    (1986)
  • R.W. Engle et al.

    Working memory, short-term memory, and general fluid intelligence: A latent-variable approach

    Journal of Experimental Psychology: General

    (1999)
  • S.E. Gathercole et al.

    The structure of working memory from 4 to 15 years of age

    Developmental Psychology

    (2004)
  • Cited by (100)

    • Working memory development in different modalities in children and young adults

      2022, Journal of Experimental Child Psychology
      Citation Excerpt :

      Although WM is a function used throughout the life span, it has particular importance during childhood because of its association with various cognitive skills that are an undifferentiated part of development such as learning, attention, mathematics, and language comprehension (Alloway et al., 2005; Archibald, 2017; Gathercole, 1999; Swanson, 2017; Swanson & Alloway, 2012; Vernucci, Canet-Juric, Zamora, & Richards, 2021). In addition, WM has been found to play a part in abilities such as social functioning and social behavior (Alloway et al., 2005; McQuade, Murray-Close, Shoulberg, & Hoza, 2013). Therefore, fully understanding WM’s character during childhood and later, during adulthood, is of great importance (Best, Miller, & Jones, 2009).

    • Executive function in kindergarten and the development of behavior competence: Moderating role of positive parenting practices

      2022, Early Childhood Research Quarterly
      Citation Excerpt :

      Conversely, other experts explain that elevated inhibitory control may contribute to children being inflexible and rigid, which can exacerbate internalizing problems like anxiety, as shown in several studies (e.g., Thorell, Bohlin, & Rydell, 2004). Poor working memory has also been linked to externalizing behaviors (Flouri, Ruddy, & Midouhas, 2017) as deficits may limit children's ability to retrieve social goals and effectively navigate socially challenging situations, leading to increased aggression (McQuade, Murray-Close, Shoulberg, & Hoza, 2013). Additionally, difficulty with working memory is often associated with child internalizing behaviors (Flouri et al., 2017), as it may impede their ability to retrieve and focus on positive rather than negative thought patterns, increasing risk for preservation and rumination.

    View all citing articles on Scopus
    View full text