Temperament factor structure in fragile X syndrome: The Children's Behavior Questionnaire

https://doi.org/10.1016/j.ridd.2013.11.024Get rights and content

Highlights

  • Distinct patterns of temperament in fragile X syndrome (FXS) are associated with clinical outcomes.

  • The factor structure the Children's Behavior Questionnaire is largely retained in children with FXS.

  • Different CBQ factor loadings in FXS may reflect syndrome-specific self-regulation, fear, and shyness.

Abstract

Early patterns of temperament lay the foundation for a variety of developmental constructs such as self-regulation, psychopathology, and resilience. Children with fragile X syndrome (FXS) display unique patterns of temperament compared to age-matched clinical and non-clinical samples, and early patterns of temperament have been associated with later anxiety in this population. Despite these unique patterns in FXS and recent reports of atypical factor structure of temperament questionnaires in Williams Syndrome (Leyfer, John, Woodruff-Borden, & Mervis, 2012), no studies have examined the latent factor structure of temperament scales in FXS to ensure measurement validity in this sample. The present study used confirmatory factor analysis to examine the factor structure of a well-validated parent-reported temperament questionnaire, the Children's Behavior Questionnaire (Rothbart, Ahadi, Hershey, & Fisher, 2001), in a sample of 90 males with FXS ages 3–9 years. Our data produced a similar, but not identical, three-factor model that retained the original CBQ factors of negative affectivity, effortful control, and extraversion/surgency. In particular, our FXS sample demonstrated stronger factor loadings for fear and shyness than previously reported loadings in non-clinical samples, consistent with reports of poor social approach and elevated anxiety in this population. Although the original factor structure of the Children's Behavior Questionnaire is largely retained in children with FXS, differences in factor loading magnitudes may reflect phenotypic characteristics of the syndrome. These findings may inform future developmental and translational research efforts.

Introduction

For decades, researchers have studied temperament as a means to investigate the biobehavioral processes underlying individual differences. Rothbart and Derryberry (1981) describe temperament as relatively stable individual differences in reactivity and self-regulation; processes influenced by genetics, experience and development. The psychosocial relationships of temperament to cognitive and behavioral outcomes are believed to be rooted in neurobiological development reflecting gene–brain–behavior relationships, particularly in prefrontal and limbic regions of the brain (see Whittle, Allen, Lubman & Yücel, 2006, for review). From this psychobiological perspective, the study of temperament involves examining and appreciating strengths and limitations of an individual's response to the environment. Moderate variability in temperament patterns is expected across individuals and reflects complex interactions of biological, environmental, and experiential factors that result in a good or poor “fit” in terms of the match between temperament traits and environmental demands. Extreme patterns of temperament often reflect maladaptive reactivity or regulatory abilities and have been linked to a host of pathological outcomes including conduct problems, hyperactivity, anxiety disorders and withdrawal (Rothbart and Bates, 2006, Rothbart et al., 1995). Thus, the study of temperament “allows for a view of the tremendous diversity and individual integrity of functioning existing even from the earliest months of life” (Rothbart & Derryberry, 1981, p. 39).

Rothbart et al. (2001) describe fifteen clusters of temperament characteristics (Table 1) that may be measured over time using developmentally specific versions of parent-rated temperament measures. These measures differentiate infant, toddler and early childhood as distinct developmental stages in which temperament characteristics vary in terms of their emergence and the quality of behaviors that vary over time. For example, negative affect and surgency are present in early infancy; however, infants present reactive affiliation and orienting characteristics that mature into self-regulatory effortful control during toddlerhood (Putnam, Rothbart, & Gartstein, 2008).

In contrast to earlier work that focused on the individual temperament subscales, recent research has examined the latent structure and interrelationship among these temperament subscales in greater detail. Several factor analytic studies of typically developing children using temperament measures have consistently yielded a three-factor structure consisting of negative affectivity, surgency (also termed “extraversion”), and effortful control (Rothbart, Ahadi, & Evans, 2000). These factors are generally stable over time. Negative affect reflects negative emotional reactivity and includes constructs such as fear, approach, soothability, sadness, anger, discomfort, and motor activity (Putnam et al., 2008). Surgency characterizes a child's desire for warmth and closeness with others and is reflected in both positive affective responsiveness to stimulation and novelty and speed of response and level of gross motor activity (Rothbart, 2007). Effortful control includes the capacity to focus and maintain short attention, control or inhibit impulses, and to derive pleasure from low-intensity stimuli (Kochanska, Murray, & Harlan, 2000). These factors have been increasingly conceptualized as interrelating risk and protective factors for psychosocial development (e.g. Fox et al., 2001, Putnam et al., 2008, Rothbart, 2007).

In light of the well-established relationship between temperament and psychosocial outcomes in typically developing children, a growing body of work has begun examining the relationship between temperament and atypical development in clinical pediatric samples. Children with Williams syndrome, a neurodevelopmental disorder caused by the deletion of ∼25 genes on chromosome 7q11.23 (Klein-Tasman and Mervis, 2003, Tomc et al., 1990). In the only published study to investigate the factor structure of temperament in neurodevelopmental disorders that we are aware of, Leyfer et al. (2012) reported a four-factor model of temperament that differentiated children with Williams syndrome from typically developing norms. The four-factor solution for Williams syndrome included constructs from both infant and early childhood stages, potentially reflecting the delayed developmental changes observed in Williams (Leyfer et al., 2012). Interestingly, the authors also identified differential associations between specific temperament factors and diagnoses of anxiety and attention deficit hyperactivity disorder in this sample.

In addition to this work in Williams syndrome, several studies have examined atypical patterns of temperament in young children with fragile X syndrome. Fragile X syndrome (FXS) is the most common inherited form of intellectual disability and is caused by an unstable expansion of the CGG repeat sequence at Xq27.3 (>200 CGG repeats). This expansion reduces FMR1 protein production, which dysregulates messenger RNA translation and impairs brain function (Bassell & Warren, 2008). Fragile X syndrome affects approximately 1:4000 males and 1:6000 females (Kaufmann & Moser, 2000), with females often experiencing less severe symptoms. Because the molecular mechanisms of FXS are well characterized, FXS is a robust model for investigating the interplay of genes, environment, behavior, and neurobiology. The cognitive and behavioral phenotype often includes hyperactivity, gaze aversion, social withdrawal and anxiety, impulsivity, aggression, stereotypic behaviors, and autism (Hall, Burns, Lightbody, & Reiss, 2008; Roberts et al., 2009, Woodcock et al., 2009). The majority of persons with FXS also meet criteria for comorbid conditions including anxiety (86%, Cordeiro, Ballinger, Hagerman, & Hessl, 2011) and attention deficit hyperactivity disorder (90%, Hagerman & Hagerman, 2002).

A number of studies have compared temperament in FXS and other clinical and non-clinical groups using experimental, physiological and parent-reported dimensions of temperament. This work indicates children with FXS are rated as more active; as well as less adaptable, intense, sad, angry, persistent, and approachable compared to typically developing peers (Hatton et al., 1999, Shanahan et al., 2008). Temperament profiles have also been used to differentiate boys with FXS from same-age peers with autism (Bailey, Hatton, Mesibov, Ament, & Skinner, 2000) and developmental delay (Kau, Reider, Payne, Meyer, & Freund, 2000). These phenotypic differences are likely rooted in well-documented neurobiological self-regulation deficits associated with FXS (e.g. Hall et al., 2009, Heilman et al., 2011), supported by evidence that parent-reported temperament is associated with physiological arousal in young boys with FXS (Roberts, Boccia, Hatton, Skinner, & Sideris, 2006). Commensurate with this association between neurobiology and temperament, as well as with the well-documented association between temperament and psychopathology in nonclinical samples (e.g. Fox et al., 2001), temperament ratings have been associated with autism symptoms (Shanahan et al., 2008) and anxiety (Tonnsen, Malone, Hatton & Roberts, 2013) in young males with FXS.

To date, these studies of temperament in FXS have primarily examined temperament at the subscale level and generally assumed that parent-report scales developed in non-clinical samples can be similarly applied in FXS. However, FXS is associated with atypical patterns of cognitive and self-regulatory mechanisms that may alter the expression and interrelationships of temperament factors. Indeed, recent evidence indicates that the factor structure of temperament differs in children with Williams syndrome compared to that reported for typically developing children (Leyfer et al., 2012). In light of these findings, the present study seeks to clarify whether the latent structure of temperament in FXS parallels the documented three-prong structure (negative affect, surgency, and effortful control) present in non-clinical pediatric samples. Given findings of differentiation of children with FXS to typical controls and recent evidence on the differing factor structure in Williams syndrome, we hypothesized that the original factor structure of the Child Behavior Questionnaire in FXS would not be retained in our clinical sample.

Section snippets

Participants

Participants included 90 males with FXS from a series of longitudinal studies out of the University of North Carolina investigating the developmental trajectories of children with FXS. Participants for this study were selected between 3 and 9 years (36 and 118 months, M = 5.85 years, SD = 1.47 years) with data from one parent-rated temperament questionnaire. Eighty-four percent of the participants were White, 9% African American, 1% Hispanic, and 6% other. Each parent completed the Vineland

Data analytic procedures

The full data set contained 187 observations with 90 participants. On average, participants had two observation points at varying ages; however, some participants had only one (N = 24), some had two (N = 41), and some had three observations (N = 27). The original data analysis plan was to analyze our full sample of participants (N = 187) using a multilevel confirmatory factor analysis strategy described by Muthén and Satorra (1995). However, due to convergence problems – likely related to our sample

Discussion

Temperament research offers a model for understanding individual differences in human behavior by recognizing the interplay between heritable biologically based traits and the environment over time. As such, studying temperament offers us the opportunity to examine contributions at the neural and genetic level, as well as at the psychological level (Posner & Rothbart, 2007). Temperament research has documented the important role that individual differences play in child development, with

Acknowledgements

This study was supported by the National Institute of Child Health and Human Development (P30-HD003110-35S1), the National Institute of Mental Health (R01-MH090194-01A1 and F31-MH095318-01A1), and the Office of Special Education Programs, U.S. Department of Education (H324C990042).

References (37)

  • A.S. Kau et al.

    Early behavior signs of psychiatric phenotypes in fragile X syndrome

    American Journal on Mental Retardation

    (2000)
  • W.E. Kaufmann et al.

    Dendritic anomalies in disorders associated with mental retardation

    Cerebral Cortex

    (2000)
  • B.P. Klein-Tasman et al.

    Distinctive personality characteristics of 8-, 9-, and 10-year-olds with Williams syndrome

    Developmental Neuropsychology

    (2003)
  • R.B. Kline

    Principles and practice of structural equation modeling

    (2011)
  • G. Kochanska et al.

    Effortful control in early childhood: Continuity and change, antecedents, and implications for social development

    Developmental Psychology

    (2000)
  • O. Leyfer et al.

    Factor structure of the Children's Behavior Questionnaire in children with Williams syndrome

    Journal of Autism and Developmental Disorders

    (2012)
  • J.C. Loehlin

    Path models in factor, path, and structural equation analysis

    Latent variable models

    (2004)
  • P. Muris et al.

    The role of temperament in the etiology of child psychopathology

    Clinical Child and Family Psychology Review

    (2005)
  • Cited by (13)

    View all citing articles on Scopus
    View full text