Research in Autism Spectrum Disorder&

Research in Autism Spectrum Disorders 6 (2012) 871–880

Contents lists available at SciVerse ScienceDirect

Research in Autism Spectrum Disorders

Journal homepage: ht tp : / /ees .e lsev ier .com/RASD/defau l t .asp

An examination of the relationship between communication and socialization deficits in children with autism and PDD-NOS

Megan A. Hattier, Johnny L. Matson *

Louisiana State University, United States

A R T I C L E I N F O A B S T R A C T

Article history: Received 7 August 2011

Accepted 15 November 2011

Keywords: Autism

BISCUIT

Battelle Developmental Inventory

Communication

Infants

Autism Spectrum Disorders (ASDs) are characterized by pervasive impairments in

repetitive behaviors or interests, communication, and socialization. As the onset of these

features occurs at a very young age, early detection is of the utmost importance. In an

attempt to better clarify the behavioral presentation of communication and socialization

deficits to aid in early assessment and intervention, impairments in these areas were

examined among infants and toddlers (17–36 months) with Autistic Disorder (AD),

Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS), and non-ASD

related developmental delay. The Baby and Infant Screen for Children with aUtIsm Traits- Part1 (BISCUIT-Part1) and the Battelle Developmental Inventory, 2nd Edition (BDI-2) were utilized to examine communication and socialization levels, respectively, among these

groups. All groups significantly differed on level of socialization impairment with the

Autism group displaying the greatest impairment and the non-ASD related developmental

delay group evincing the least impairment. In regards to communication deficits, the non-

ASD related developmentally delayed group differed significantly in comparison to the

Autism and PDD-NOS groups; however, no significant differences were found between

children with AD and PDD-NOS. While communication and socialization impairments

were found to significantly correlate for all participants with the exception of those with

PDD-NOS, these correlations were not found to significantly differ from one another across

groups. The implications, limitations, and future directions of these results are discussed.

? 2011 Elsevier Ltd. All rights reserved.

In the past decade there has been an increase in the public’s interest in Autistic Disorder (AD), more commonly known as autism (Ban Itzchak, Lahat, & Zachor, 2011; Evans et al., 2001; Levy & Perry, 2011; Lord & Luyster, 2006; Matson, Wilkins, & Gonzales, 2008; Suzuki, 2011; Worley, Matson, Sipes, & Kozlowski, 2011). Autism is a neurodevelopmental disorder characterized by pervasive deficits in socialization and communication, as well as the presence of repetitive or restricted behaviors or interests (Horovitz & Matson, 2010; Lugnega?rd, Hallerba?ck, & Gillberg, 2011; Matson, 1994, 2008; Matson, Dempsey, & Fodstad, 2009; Matson, Dempsey, & LoVullo, 2009; Matson, Fodstad, Hess, & Neal, 2009; Meindl & Cannella- Malone, 2011; Worley & Matson, 2011). For purposes of this study, the focus will remain on the former two impairments. Approximately, 25–50% of all children with an ASD diagnosis never develop a functional language (Dawson & Murias, 2009; Howlin, 2006; Rutter, 1978). Many are stigmatized socially as well, particularly those who exhibit stereotypic behavior (Cunningham & Schreibman, 2008; Matson, Shoemaker, et al., 2011; Matson, Kozlowski, et al., 2011; Rivet & Matson, 2011; Smith & Matson, 2010a). Therefore, optimizing skills in communication and socialization is of great importance (Matson, Kozlowski, et al., 2011; Matson, Sipes, et al., 2011).

* Corresponding author at: Department of Psychology, Louisiana State University, Baton Rouge, LA 70803, United States.

E-mail address: [email protected] (J.L. Matson).

1750-9467/$ – see front matter ? 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.rasd.2011.12.001

http://dx.doi.org/10.1016/j.rasd.2011.12.001
mailto:[email protected]
http://www.sciencedirect.com/science/journal/17509467
http://dx.doi.org/10.1016/j.rasd.2011.12.001
http://ees.elsevier.com/RASD/default.asp
M.A. Hattier, J.L. Matson / Research in Autism Spectrum Disorders 6 (2012) 871–880872

Deficits in the area of socialization can be detrimental to a child’s quality of life for many reasons, as these children tend to isolate themselves from others and have impaired social relationships (Chan, Hu, Cui, Wang, & McAlonan, 2011; Ghuman, Leone, Lecavalier, & Landa, 2011; Mahan & Matson, 2011b; Matson, Fodstad, & Rivet, 2009; Matson, Matson, & Rivet, 2007; Smith & Matson, 2010c). The three most explicit social impairments of those with ASDs identified by Rutter (1978) include uncooperativeness while playing with other children, the inability to form friendships, and the failure to recognize others’ feelings. Communication deficits also lead to negative consequences, such as elevated levels of problem behaviors (Barnes, Dunning, & Rehfeldt, 2011; Beitchman, 2006; Matson & LoVullo, 2008; Smith & Matson, 2010a, 2010b, 2010c; Sturmey, Laud, Cooper, Matson, & Fodstad, 2010a, 2010b). More importantly, Newborg (2005) hypothesizes that children with higher deficits in communication may also exhibit greater socialization deficits, particularly, because the inability to communicate with adults and/or peers can create social strain (Matson, Fodstad, et al., 2009; Matson & Wilkins, 2007; Matson, Wilkins, & Gonzales, 2008).

Over the past 30 years there reportedly has been a 16-fold increase in the diagnosis of autism, although this increase may not be completely attributed to a genuine growth in the disorder. Rather, other factors like greater diagnostic precision, more expansive diagnostic criteria, and more public attention to the disorder may be potential causes (Bertoglio & Hendren, 2009; Matson & Kozlowski, 2011). Nevertheless, more children are in need of services and accommodations, thus making early detection and intervention a top priority among today’s diagnosticians (Matson & Boisjoli, 2007; Matson, Dixon, & Matson, 2005; Volkmar & Pauls, 2003). While much research has been conducted on the presence of communication and socialization impairments in this population, the amount of literature is much less for younger children, especially for children under the age of 3. This factor may be attributed to the fact that the average age for diagnosis of ASDs is approximately 3–4 years of age, although it is dropping (DeGiacomo & Fombonne, 1998; Matson, 2005). The current study serves as a means to provide such an analysis as all participants were 17–36 months of age. The majority of studies which have been conducted to analyze core ASD symptomatology among this very young population have done so through the utilization of retrospective analyses (e.g., inspecting old home videos; Brown, Dawson, Osterling, & Dinno, 1998; Osterling & Dawson, 1994). The current study is more robust since deficits in communication and socialization were examined using real time, objective measures.

In 2010, Horovitz and Matson found children 17 to 36 months of age with PDD-NOS possess significantly more communication deficits than those with non-ASD related developmental delays. Children with a diagnosis of Autistic Disorder were found to display significantly more deficits in communicative skills than children with PDD-NOS and non-ASD related developmental delay. The current study extends the Horovitz and Matson (2010) study and adds additional children to the sample. The updated sample included newly recruited participants and only children that have been administered the Battelle Developmental Inventory, 2nd Edition (BDI-2). Additionally, the present study examines not only the presence of communication deficits but the presence of socialization impairments and the relationship of the deficits seen in these two areas.

Since Kanner’s (1943, 1944) original description, many researchers have conducted studies in an attempt to better define the three core features of ASD on an individual basis (Kanai et al., 2011; Sipes, Matson, Worley, & Kozlowski, 2011; Tseng, Fu, Cermak, Lu, & Shieh, 2011). However, understanding the relationship between the core features of autism is equally important when attempting to detect them among young children. For a diagnosis of AD, all three core features must be present to a large degree, which suggests there are strong associations between these symptoms (Dworzynski, Happe?, Bolton, & Ronald, 2009; Kuenssberg & McKenzie, 2011; Matson, Matson, & Beighley, 2011). Some, however, have questioned these associations. For instance, family studies have found that relatives of those with autism often display milder forms of communication and socialization deficits without having repetitive and restricted behaviors and interests (Bishop et al., 2004; Piven, Palmer, Jacobi, Childress, & Arndt, 1997). More recently, Dworzynski et al. (2007) found significant correlations between communication and socialization impairments among children of this ASD population; however, there was no significant correlation between communication impairment and repetitive and restricted behaviors and interests. Elsewhere, Howlin and Moore (1997) state that communication and socialization deficits are the first signs suggesting a child is developing atypically.

The aim of the current study was to build upon these abovementioned findings. Fortunately, due to recently developed instruments, screening for autism and other developmental delays has proved to be less problematic than in the past. The Baby and Infant Screen for Children with aUtIsm Traits-Part1 (BISCUIT-Part1) has recently been designed to aid in the early detection of ASDs among children from 17 to 37 months of age (Matson, Wilkins, Sevin, et al., 2008). Conversely, the BDI-2 (Newborg, 2005) is intended to identify developmental skills of children from birth to 7 years 11 months. This study aimed to utilize two portions of these two measures (the communication domain of the BISCUIT-Part 1 and the Personal-Social domain of the BDI-2) in establishing if a relationship exists, and if so determining where the correlation lies between the level of communication deficits and the level of socialization deficits among those with diagnoses of AD, PDD-NOS, and non-ASD related developmental delays.

First, it was hypothesized that the autism group would have significantly greater levels of impairment in communication and socialization than the PDD-NOS and non-ASD related developmental delay groups. Those with PDD-NOS were also expected to show significantly greater levels of deficits in these areas compared to the non-ASD related developmental delay group. Second, it was also thought that correlations between level of communication deficit and socialization impairment would be significant for the AD group; however, it was believed that non-significant differences would be found for the PDD- NOS and non-ASD related developmental delay groups. Finally, in comparison of these correlations for each diagnostic group, it was hypothesized that significant differences would be found between the communication–socialization (C–S)

M.A. Hattier, J.L. Matson / Research in Autism Spectrum Disorders 6 (2012) 871–880 873

correlations for those with autism and those with PDD-NOS and for those with autism and those with non-ASD related developmental delay. It was not believed, however, that the C–S correlations would significantly differ for the PDD-NOS group and the non-ASD related developmental delay group.

1. Method

1.1. Participants

Five hundred ninety-one children served as participants. Ranging from 17 to 37 months of age (M = 26.03; SD = 4.71) these infants and toddlers were recruited through the Early Steps program funded by the State of Louisiana. Early Steps is Louisiana’s Early Intervention System housed under the Individuals with Disabilities Education Act, Part C. Infants and toddlers from birth to 36 months of age who had developmental delays or a medical condition likely to result in a developmental delay qualify for services. Participants were classified into one of these three conditions: Autism, PDD-NOS, or controls with non-ASD related developmental delay. These assignments were established by a licensed doctoral level psychologist who was blind to the BISCUIT scores (Matson, Boisjoli, Hess, & Wilkins, 2010). A portion of participants from the original sample recruited for this study (n = 197) also received diagnoses from a second doctoral level clinical psychologist, and inter-rater reliability between these independent diagnoses was found to be excellent (k = .935). The non-ASD related developmental delay group consisted of children who did not meet criteria for an ASD but their previous family pediatrician determined that they were either atypically developing, had a genetic disorder, or had a physical disability (Matson, Fodstad, et al., 2009).

Originally, a total of 2214 participants were recruited. All participants with missing or improperly coded data were removed from inclusion in this study (n = 818). Consequently, the PDD-NOS group was the smallest diagnostic group with 197 participants. Field (2009) suggests generating equal sample sizes among all groups to ensure robustness of statistical tests; thus, all three diagnostic groups were made equal, each with 197 participants. This process was conducted by utilizing the select random cases function in SPSS. Therefore, one participant was randomly deleted from the AD group and 804 participants were deleted from the non-ASD related developmentally delayed group to achieve group totals of 197 for each group.

The children within the autism group ranged in age from 18 to 36 months of age (M = 26.59; SD = 4.75). For this diagnostic group, 47.2% were Caucasian, 43.1% were of African American descent, 2.5% were of Hispanic ethnicity, and 7.1% were recorded as other. Additionally, 75.1% of the autism group was male. In regards to the PDD-NOS group, children from 17 to 35 months of age (M = 25.54; SD = 4.48) met inclusion criteria. The ethnicities of these children with a PDD-NOS diagnosis were recorded as Caucasian (48.7%), African American (44.7%), Hispanic (.5%), or other (6.1%). Within this group, 72.6% were male. Within the non-ASD related developmental delay group, the children were between the ages of 18 and 36 months (M = 25.96; SD = 4.86) with 66% being male. In regards to ethnicity, 51.3% were Caucasian, 44.2% were African American, 1.5% were Hispanic, and 3.0% were recorded as other. Demographic information is presented within Table 1.

To determine if the diagnostic groups differed significantly on demographic variables (i.e., gender, ethnicity, or age) a priori analyses were conducted (Matson, Rivet, Fodstad, Dempsey, & Boisjoli, 2009). The results from chi-square analyses revealed that the groups did not differ significantly in terms of gender or ethnicity. A one-way between-subjects analysis of variance (ANOVA) also found no significant differences between groups in terms of age. While non-significant differences among these variables exist, it is believed that this variability would not significantly affect the findings from this study.

1.2. Measures

1.2.1. Baby and infant screen for children with autism traits – Part 1

The BISCUIT has recently been designed to aid in the early detection of ASDs among children from 17 to 37 months of age (Matson et al., 2010; Matson, Wilkins, Sevin, et al., 2008). It is a battery of assessments designed to assess autism in young

Table 1

Demographic characteristics (N = 591).

Demographic characteristics Diagnostic group

Autism (n = 197) PDD-NOS (n = 197) Non-ASD Developmentally Delayed (n = 197)

Age (in months)

Mean (SD) 26.59 (4.75) 25.54 (4.48) 25.96 (4.86)

Range 18–36 17–35 18–36

Gender, %

Male 75.1% 72.6% 66.0%

Female 24.9% 27.4% 34.0%

Race/ethnicity, %

Caucasian 47.2% 48.7% 51.3%

African–American 43.1% 44.7% 44.2%

Hispanic 2.5% 0.5% 1.5%

Other 7.1% 6.1% 3.0%

M.A. Hattier, J.L. Matson / Research in Autism Spectrum Disorders 6 (2012) 871–880874

children along with PDD-NOS, comorbid psychopathology, and challenging behaviors. The BISCUIT-Part 1 is the section concerned with diagnostic criteria and consists of 62 questions. The parents and/or caregivers rate the child’s impairments in comparison to typically developing children of the same age. Items are scored on a 3-point scale: 0 indicating no difference or no impairment; 1 indicating different or mild impairment; and 2 indicating very different or severe impairment in comparison to their peers. Inspection of these items with a factor analysis revealed three separate factors: socialization/ nonverbal communication, repetitive behaviors/restricted interest and communication (Matson et al., 2010). The seven items that fall under the communication factor will be the focus of this study. These items include ‘‘use of language to communicate,’’ ‘‘language development,’’ and ‘‘communicates effectively.’’ This communication domain has been determined to have good internal consistency (0.83) and item-scale correlations ranging from .34 to .90 (Matson et al., 2010).

Internal reliability for this 62 question component was found to be high .97 (Matson, Wilkins, Sevin, et al., 2008). Item content for autism and PDD-NOS was successfully established. Validity studies found that the BISCUIT-Part 1 was able to effectively distinguish between those with and without ASDs. Furthermore, as previously mentioned when differentiating between those without a diagnoses and those with PDD-NOS the sensitivity and specificity was established as .847 and .864, respectively (Matson, Wilkins, et al., 2009). The sensitivity and specificity was found to be slightly lower (.844 and .833, respectively) when distinguishing between diagnoses of autism and PDD-NOS (Matson, Wilkins, et al., 2009). Lastly, the overall classification rate was found to be 88.8 for the BISCUIT-Part 1 (Matson, Wilkins, et al., 2009).

1.2.2. Battelle developmental inventory, second edition

The BDI-2 (Newborg, 2005) is a revision of the original BDI. The revisions include, but are not limited to, relocating the placement of some items into different domains, fewer subtrials on many items for efficiency purposes, and an easier-to- administer design of the interview. It is intended to identify developmental skills of children from birth to 7 years 11 months. Administration of the full BDI-2 usually lasts approximately 1–2 h. The five domains that this 450-item assessment addresses are Adaptive (ADP), Personal–Social (P–S), Communication (COM), Motor (MOT), and Cognitive (COG). The items are scored on a 3-point Likert scale: a score of 0 indicates that the child has no ability in this skill; a score of 1 indicates that they possess an emerging ability; and a score of 2 indicates that they have ability with this skill. A total developmental quotient (DQ) is calculated by combining the scores of each of the five domains. This combined score has a mean of 100 and a standard deviation of 15. Using a sample of 2500 children between the ages of birth to 7 years, 11 months, acceptable levels of test retest reliability and excellent internal consistency were found along with appropriate content and criterion validity (Newborg, 2005).

For the purposes of this study, the score for the Personal–Social domain was used as the dependent variable. This particular domain consists of 100 items that assess the child’s ability to interact with adults and peers and their self-concept and self role (Newborg, 2005). The test is made up of 3 subdomains: Adult Interaction (AI), Peer Interaction (PI), and Self- Concept and Social Role (SR). The AI subdomain includes 30 items that are administered to only children younger than 6 years of age. Attachment to and interaction with adults during infancy is assessed along with initiation and maintenance of social contact and the use of adults to assist themselves with solving problems. The assessment of the 25-item PI subdomain begins at 2 years of age and ends at 6 years of age. Behaviors including, but not limited to, forming friendships, interacting with peers, responding to and initiating social contact with peers, playing well in a small group, and cooperation are among the abilities assessed in this subdomain. The SR subdomain consists of 45 items that are administered to all ages that the BDI- 2 assesses (birth to 7 years, 11 months). The child’s self-awareness, self-worth, morals, sensitivity to the feelings of others, and coping skills are among the skills addressed in this subdomain. The quality and frequency of the abilities mentioned above are also measured for each item on each subdomain.

Good internal reliability has been found for the BDI-2 total score and for the Personal–Social Domain, .99 and .96, respectively (Newborg, 2005). All subdomains within the P–S domain also reached adequate levels in regards to internal reliability. The test–retest was calculated using a sample of 126 two-year-old children. The stability coefficient for the P-S domain and the total DQ were both very high, .90 and .93, respectively (Newborg, 2005). Inter-rater reliability was also found to be quite high. Consistency between scorers ranged from 94% to 99% agreement (Newborg, 2005). Convergent validity was also established with a number of different scales measuring development in young children. BDI-2 scores were also able to effectively distinguish between typically developing children with children with autism evidenced by sensitivity and specificity levels of .91 (Newborg, 2005).

1.3. Procedure

Parental interviews and child observations were conducted by individuals whose training qualified them to screen children that might benefit from services provided by EarlySteps. Physical therapy, occupational therapy, social work, education, speech–language pathology, and psychology were the various disciplines represented (Matson, Wilkins, et al., 2009). In addition to their prior training, the assessors received education on ASDs, the measures used throughout the screening process, and the correct standardized administration methods. The screening process involves an entire battery of assessments which include the BISCUIT and the BDI-2. The parents or legal guardians of the children participating in this study served as the informants on all measures and have provided informed consent for participation. Furthermore, the Louisiana State University Institutional Review Board and Louisiana’s Office for Citizens with Developmental Disabilities provided prior approval for this study.

M.A. Hattier, J.L. Matson / Research in Autism Spectrum Disorders 6 (2012) 871–880 875

2. Results

2.1. Main analyses

All statistical analyses were carried out using SPSS 16.0. A 3? 2 MANOVA was conducted to test for differences among the three groups (Autism, PDD-NOS, and non-ASD related developmental delay) on communication and socialization scores (i.e., the overall communication score on the BISCUIT-Part 1 and the developmental quotient on the P–S domain of the BDI-2, respectively). Determined to be robust in cases of equal sample sizes (Field, 2009), the Pillai–Bartlett trace statistic indicated significant differences between the three diagnostic groups in regards to communication and socialization scores, F(4, 1176) = 71.69, p< .001. Significant results of the MANOVA were followed by 2 one-way between subjects ANOVAs to determine if these significant group differences lie among the P–S domain score of the BDI-2 or the overall communication score of the BISCUIT- Part 1. Again, the diagnostic groups served as the independent variable. Separate univariate ANOVAs on the outcome variables were significant across diagnostic group effects on communication, F(2, 588) = 117.22, p< .001, and socialization, F(2, 588) = 86.08, p< .001. To account for the inflation of type-I errors, Tukey post hoc tests were conducted following significant ANOVAs. In regard to communication, the Autism group did not differ significantly when compared to the group with PDD-NOS, p = .062. Significant differences were found for communication, however, between the Autism group and the non-ASD related developmental delay group, p< .001, and between the PDD-NOS group and the non-ASD related developmental delay group, p< .001. See Fig. 1 for a depiction of the mean scores on the communication domain of the BISCUIT-Part 1 for all diagnostic groups. For socialization, all comparisons (i.e., Autism vs. PDD-NOS; Autism vs. non-ASD related developmental delay; and PDD- NOS vs. non-ASD related developmental delay) were found to differ significantly, p< .001, universally. See Fig. 2 for a depiction of the mean scores for children with autism, PDD-NOS, and non-ASD related developmental delay on the P–S domain of the BDI-2. The next set of analyses involved identifying the relationship between communication and … Read more Applied Sciences Architecture and Design Biology Business & Finance Chemistry Computer Science Geography Geology Education Engineering English Environmental science Spanish Government History Human Resource Management Information Systems Law Literature Mathematics Nursing Physics Political Science Psychology Reading Science Social Science Home Homework Answers Blog Archive Tags Reviews Contact google+twitterfacebook Copyright © 2021 HomeworkMarket.com