Social Competence and its Relationship with the Neuropsychological Profile in Children with Autism Spectrum Disorder and Children with Developmental Language Disorder

(1)

Social Competence and its Relationship with the

Neuropsychological Profile in Children with

Autism Spectrum Disorder and Children with

Developmental Language Disorder

Ruth Peeters

30the June 2019

MA Thesis in Neurolinguistics

Departments of Language and Cognition

Faculty of Arts

University of Groningen

Supervisors

Dr. W. Tops

Prof. dr. B. Maassen

(2)

(3)

Abstract

Introduction Children with Autism Spectrum Disorders (ASD) and children with

Developmental Language Disorders (DLD) show an overlap in symptoms, in particular both groups show social impairments. Therefore the diagnostic boundaries between those two neurodevelopmental disorders are not clear-cut. In the present study it is supposed that there is variability in the underlying nature driving these social problems between both groups. For this reason, the neuropsychological bases of social impairment in ASD and DLD are investigated. The possible overlap in the social profile of ASD and DLD and the suggestion that their common social impairments comes from different causes will be explored by: (1) Identifying whether children with ASD carry different cognitive characteristics than the children with DLD.

(2) Determining whether their social communication skills are related to their cognitive abilities.

Method Data were collected from a database on archival research of a multidisciplinary

centrum for children with neurodevelopmental disorders. The data of 75 children with ASD and 26 children with DLD were used in this study. With the data of a cross battery assessment of a broad range of neuropsychological functions, a Repeated Measures Mixed Models analysis is performed to evaluate the strengths and weaknesses within the profile of both groups and to compare between them. The social functioning of each group is evaluated based on the scores of the Social Responsiveness Scale (SRS). These scores are compared within and between both groups, also using a Repeated Measures Mixed Models. Finally, correlations between the neuropsychological component scores and SRS scores have been examined for both the group with ASD and the group with DLD.

Results The interaction between neuropsychological components and group emerged from

the Repeated Measures Mixed Model, and was significant. This indicates that the profile of between the neuropsychological components did differentiate between the two groups. The group with ASD showed other strengths and weaknesses in their neuropsychological functioning than the children with DLD. The children with ASD showed higher scores on Visual Processing and Comprehension in comparison to Processing Speed and Verbal Working Memory. In the group of children with DLD higher scores on Fluent Reasoning, Visual Processing and Processing Speed were obtained in relation to their scores on Verbal

(4)

Working Memory and Auditory Processing. Correlation analysis showed differences between neuropsychological components and social functioning for the group with ASD and the group with DLD.

Discussion The groups with ASD and with DLD have clearly distinctive neuropsychological

profiles. The strengths and deficits are not the same in the group of ASD and the group of DLD, and different core features drive the co-occurring social dysfunctions between both groups. Measurements of a range of broad neuropsychological functions may contribute to validate differential diagnosis. Results of the present study therefore motivate the use neuropsychological assessments (in form of a CHC-model) in diagnosing DLD and ASD in a clinical setting.

(5)

Acknowledgements

The current thesis forms the final assignment of the master in Neurolinguistics from the University of Groningen. There are numerous persons that I would like to thank for their assistance in the preparation, writing and finishing of this thesis.

I would like to thank Dr. Wim Tops for his guidance throughout this study. First of all for letting me work as intern at his multidisciplinary center in Flanders, introducing me to the joy of working with children with neurodevelopmental disorders. I want to thank him for the opportunities he has given, to gain more experience as neurolinguist and to pursue my field of interests in this experimental research. His friendly guidance and expert advice have been invaluable throughout all stages of the work.

I am very grateful for the assistance I received from Prof. dr. Ben Maassen, for his guidance, involvement and editing.

Thanks are also expressed to my colleagues at the multidisciplinary center in Flanders, for always being very welcoming, supporting and for making the data collection possible.

I owe many thanks to my parents for offering me all the help that I ever needed. In addition, particular thanks to my boyfriend Gert and my friends, without whose support, writing this thesis would have been a lot harder.

(6)

List of Abbreviations

ADI-R Autism Diagnostic Interview-Revised

ADOS Autism Diagnostic Observation Scale

ASD Autism spectrum disorder

BRIEF Behavior Rating Inventory of executive Function CELF-4-NL Clinical Evaluation of Language Fundamentals-4-NL

CMS Children Memory Scale (CMS; Cohen,

DLD Developmental Language Disorder

DSM-5 Diagnostic and Statistical Manual of Mental Disorders 5th edition

FDI Freedom from Distractibility Index

MNS Mirror Neuron System

PIQ Performance Intelligence Quotient

POI Perceptual Organization Index

PSI Processing Speed Index

PSTM Phonological Short Term Memory

SON-R Snijders-Oomen Nonverbal Intelligence Test

SPCD Social Pragmatic Communication Disorder

ToM Theory of Mind

VCI Verbal Comprehension Index

VIQ Verbal Intelligence Quotient

WISC-III Wechsler Intelligence Scale for Children 3th edition

(9)

1. Introduction

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by persistent deficits in social communication and interaction, in combination with restricted, repetitive patterns of behavior, interests, or activities (American Psychiatric Association, 2013). The term ASD covers a large spectrum of symptoms, skills, and levels of impairment (Brentani et al., 2013). The presentation of ASD ranges widely, from a child who is nonverbal and has more difficulty with social interactions, to a child who is verbally fluent, but overly reliant on previously learned scripts of speech and social behavior. For this reason, a significant minority of children with ASD is likely to be undiagnosed (Huerta & Lord, 2012). Weaknesses in social skills are universal in children with ASD across ages and ability levels, in spite of heterogeneity of language abilities (Tager-Flusberg, Joseph, & Folstein, 2001). Although impairments in social communication are a hallmark of ASD, relatively little is known about the underlying mechanisms of these deficits. Furthermore, the deficits in social communication in ASD are also observed in individuals with Developmental Language Disorder (DLD) (Cantwell, Baker, Rutter, & Mawhood 1989; Howlin, Mawhood, & Rutter 2000).

DLD is a neurodevelopmental disorder in which language deficits that results in problems with comprehension, production, or usage of language is seen. When a language disorder is the primary disability, with no comorbidities such as sensory impairment, intellectual disability, and motor dysfunction it is considered a developmental language disorder (American Psychiatric Association, 2013). Social skills of children with DLD are recognized to be lower than expected for their age (Norbury, Nash, Baird, & Bishop, 2004). In addition Leyfer, Tager-Flusberg, Dowd, Tomblin, and Flostein (2008) reported that 41% of a group of children with DLD met diagnostic thresholds for ASD in social and communication domains on “gold standard” measures of ASD. In contrast to children with ASD, in DLD the core difficulties are rather in structural language, that is, in phonology, morphology, and syntax. Although it is known that language difficulties have an impact upon social behavior in DLD, capturing the extent of social difficulties in this group can be challenging. Even though social difficulties exist in children with DLD, they have been described as mild and secondary to the language impairment itself.

(10)

Although there is overlap in social behavioral problems between individuals with ASD and individuals with DLD, there could be variability in the underlying mechanisms driving these problems. This is in contrast to the suggestion of Bishop (2010), that ASD and DLD are part of the same spectrum, or, as Bishop coins it, the same “continuum”. Bishop takes the above-chance co-occurrence of symptoms in ASD and DLD, as an indication that they share the same etiology. For this reason, it is important to study the extent and nature of these co-occurring social impairments between both groups. The present study examines the underlying mechanisms of social functioning in terms of the neuropsychological profile of both groups. The aim of this thesis is to understand to what extent social functioning relates to differences in the neuropsychological profile in children with ASD and with DLD.

(11)

2. Literature Review

2.1 Social Information Processing

Social communication is a complex activity relying on both linguistic skills and pragmatic competence. Proficient communication processing involves the ability to efficiently carry out many different linguistic and cognitive tasks. The rapid integration of much and varied linguistic and extralinguistic contextual information imposes complex processing demands, which are taxing on the individual’s attention, memory, mind-reading, and inferential abilities (Bara, 2010; Sperber & Wilson, 2002). Indeed, social skills include a broad array of verbal and nonverbal behaviors used in reciprocal social interaction.

Dodge (1986) revealed a theoretical approach of social information processing, which was reformulated by Crick and Dodge (1994). The reformulated model describes a circular process in which a series of stepwise mental mechanisms are activated in response to an external social cue and deactivated upon the individual’s enactment of a behavioral response. The mental steps include: (1) the encoding of social cues; (2) the interpretation of the cue; (3) clarification of goals; (4) response construction; and (5) response decision (Crick & Dodge, 1994). After these five steps the enactment of the behavioral response follows. In the initial steps, individuals have to selectively focus on particular cues and, based on these cues, interpret the context of the situation. In steps 3, 4, and 5, possible responses from previous experiences stored in long-term memory are accessed, and individuals have to select their goal for interaction and evaluate the possible outcomes of these responses. Studies hypothesized that children with ASD have problems with steps 1–3 and that children with DLD have rather problems with steps 3–5 (Adams, Lockton, & Collins, 2018). In this way, the social impairments in children with DLD are generally considered to be associated with poverty of expressive language skills and delayed comprehension development (Marton, Abramoff, & Rosenzweig, 2005). Indeed, the communicative deficits described in children with DLD, consist in particular of atypical word choices, literal interpretation of figurative language, poor topic maintenance, poor turn-taking and conversational inadequacies, but it consist of some fundamental deficits in social cognition as well, such as appreciating the thoughts and feelings of others (Ketelaars, Cuperus, Van Daal, Jansonius, & Verhoeven 2009).

(12)

These described social impairments seem to be even more prominent in the group with ASD, and furthermore, the extent of these deficits seems to differ among both groups. Persistent deficits in social communication and interaction in ASD are described in the DSM-5 criteria, as deficits in social-emotional reciprocity (e.g. reduced sharing of interests, emotions or affect, a failure to initiate or respond to social interactions), deficits in nonverbal communicative behaviors (abnormalities in eye contact and body language, deficits in understanding and use of gestures, a total lack of facial expressions) and eventually as deficits in developing, maintaining, and understanding relationships (American Psychiatric Association, 2013). The deficits in nonverbal communicative behavior are not seen in children with DLD. In the study of Mailena-Arnold, Alibali, Hostetter and Evans (2014) children with DLD produced even significantly more representational gestures than their peers with a typical development. In addition, a lack of eye contact is not specified as a symptom of DLD. This suggests that the use of nonverbal communication seems to be a natural and perhaps useful way to communicate in children with DLD.

In the next sections the social functioning in children with ASD and DLD is discussed, along with the possible underlying mechanisms of these social impairments.

2.2 Social Functioning in Children with Autism Spectrum Disorder

A lot of research has been conducted on individuals with ASD in an attempt to isolate the specific core deficits of the social deficit. Most of the research on these social cognitive difficulties has focused on theory of mind (ToM) and executive functions (e.g., Adams, Green, Gilchrist, 2002; Baron-Cohen, 1997; Baron-Cohen, Leslie, Frith 1985; Heavey, Philips, Baron-Cohen, & Rutter, 2000; Hill & Russell, 2002; Lerner, Hutchins, & Prelock 2011). Lately, the presence of social impairments of children with ASD has been explained by a dysfunction in the mirror neuron system. The three theories are discussed briefly in the next paragraphs. Subsequently the neuropsychological profile of children with ASD is discussed in the section 2.3, and forms the basis of the research of the relationship between the social functioning and neuropsychological functioning in children with ASD.

2.2.1 Impairments in Theory of Mind

Baron-Cohen et al. (1985) described as one of the first a failure to represent mental states in children with ASD, using the well-known ‘Sally–Anne task’. In this test, the child is

(13)

introduced to two characters, Sally and Anne. Sally puts a marble in a basket so that Anne can see. Anne then leaves the room, and Sally moves the marble to the box. The child is asked; “When Anne comes back, where will she look for the marble?” and most of the children with ASD reply “in the box”. The term ‘mental states’ is used to refer to knowledge, beliefs, feelings, intentions, and desires. This deficit has also been called “mind-blindness” (Baron-Cohen, 1995b). Indeed, a lot of research supports the influence of ToM on the social functioning of children with ASD (Kimhi, 2014). Mazza et al. (2017) showed that ToM plays a key role in the development of social abilities, and that the lack of ToM competences in children with ASD impairs their competent social behavior. Therefore the significance of the ToM hypothesis in the interpretation of core deficits in both social and language domains of ASD should not be underestimated. However, not all aspects of this neurodevelopmental disorder can be interpreted within this framework (Tager-Flusberg, 1999). That is why several theories attempt to explain the social deficits in autism without postulating the absence of a specialized theory-of-mind mechanism.

2.2.2 Impairments in Executive Functioning

Some of these studies highlight a primary deficit in executive functioning, considering that individuals with ASD fail to make cognitive shifts based on changes in task demands. The executive processes of behavioral regulation, including inhibition, emotional regulation, and shift are good predictors for social functioning not only in children with a typical development but also in children with high-functioning ASD. However, the metacognitive executive processes of initiation, short-term memory and planning are predictors of social adaptation only in children with ASD (Leung, Vogan, Powell, Anagnostou, & Taylor, 2016). Various reviews have suggested a specific pattern of executive dysfunction in components such as planning and flexibility that distinguishes ASD from other disorders (Wallace et al., 2016). In addition negative relationships have been found between the working memory and the ‘initiate’ subdomain of the Behavior Rating Inventory of executive Function (BRIEF; Isquith, Gioia, Guy, Kenworthy & Staff, 2015) (Gilotty, Kenworthy, Sirian, Black, & Wagner, 2002). Freeman, Locke, Fulle and Mandell (2017) also suggest that poorer executive functioning is associated with increased playground isolation and less engagement with peers.

(14)

2.2.3 Mirror-neuron System Dysfunction

Some researchers suggest that social communication impairments could be due to the dysfunction of the mirror neuron system (MNs). The MNs is a group of specialized neurons in the premotor cortex, the supplementary motor area, the primary somatosensory cortex and the inferior parietal cortex, that ‘mirrors’ actions of others. The MNs regions support various aspects of sensorimotor integration (Iacoboni & Dapretto, 2006). These neurons are involved in perception and comprehension of motor actions, but also in higher-order cognitive processes such as imitation and social communication (De Giacomo et al., 2009; Gallese, Rochat, & Berchio, 2013). A close link between motor and social impairments has therefore been observed in recent studies. For example, Fournier, Hass, Naik, Lodha, and Cauraugh (2010) suggest that motor deficits influence the development of functions, which are critical to social communication development. Also in other clinical data, poor motor skills are observed in children with ASD, and are associated with greater social difficulties (Dziuk et al., 2007; Green et al., 2009; Hirata et al., 2014).

2.2.3.1 Visual Motor Integration and Social Functioning

Visual motor integration may especially be associated with social functioning through imitation of others’ actions (Ament et al., 2015). In the study by Chien et al. (2015), lower aiming and catching abilities were correlated with greater social impairment. This difficulty with efficient visual-motor integration in children with ASD may alternatively be due to cerebellar dysfunction, as the cerebellum is crucial for sensory-motor integration (Ogawa, Inui, & Sugio, 2006; Salmi et al., 2010). Fuentes and Bastian (2007) emphasize the role of the cerebellum in two processes that could be considered aspects of motor cognition, namely the abilities to predict movement outcomes and understand the meaning of movements that contribute in different ways to both motor behavior and social behavior. Non-verbal communication relies on motor cues. For instance, hand gestures, posture, body positioning, and facial expressions are the primary tools of non-verbal interaction. The difficulties with mirroring, performing and planning physical movement in children with ASD could be associated to impairments of physical movements inherent in non-verbal communication. Next, children with ASD could not receive appropriate nonverbal responses, because they cannot model the expected nonverbal cues due to the described motor impairments. Both problems in performing and understanding non-verbal communication lead to fewer and poorer-quality interactions with peers and difficulties in understanding other’s intentions

(15)

(Cossu et al., 2012). With the afore-mentioned evidence it is reasonable to conclude that social communication and motor aspects peculiar to ASD lie within a single framework such as motor cognition, associated with impairment in mirroring.

To be brief, impairments in theory of mind, in executive functioning (especially in initiation, working memory and planning) and visual motor integration could be the foundations of a deficit in social functioning in ASD. With this in mind, the neuropsychological profile of children with ASD will be discussed, with the purpose to check if children with ASD achieve lower scores on subdomains, in which theory of mind, working memory, planning and initiation, or visual motor integration skills play a role. If so, it is interesting to see how their neuropsychological profile relates with their social communication skills.

2.3 The Neuropsychological Profile of Children with Autism Spectrum Disorder

In the study by Mouga et al. (2016) the influence of specific neurodevelopmental ASD deficits on intellectual profiles of children was investigated. The cognitive profile of children ASD measured by the WISC-III (Wechsler, 1991) was compared with that of a typical developing group of children. Mouga et al. (2016) found a significant difference between the Verbal Intelligence Quotient (VIQ) and the Performance Intelligence Quotient (PIQ) in the group of children with ASD. These results corroborate the typical VIQ-PIQ discrepancies described in individuals with ASD (Charman et al., 2011b; Ryland, Hysing, Posserud, Gillberg, & Lundervold 2014; Minshew, Goldstein, Muenz, & Payton 1992; Szatmari, Tuff, Finlayson, & Bartolucci 1990).

Further, when taking the factor analysis into consideration, it includes higher scores on the Verbal Comprehension Index (VCI) and on the Perceptual Organization Index (POI), when compared to the Freedom from Distractibility Index (FDI) and the Processing Speed Index (PSI) (Mayes & Calhoun, 2003, 2008; Nyden, Billstedt, Hjelmquist, & Gillberg 2001; Wechsler, 2003b). The FDI is an index score comprised of the sum of the scores on the Arithmetic and Digit Span subtests. The FDI is interpreted as a measure of attention and concentration and is partially replaced by the ‘working-memory index’ in the present study. The PSI measures a child’s ability efficiently to scan and quickly identify visual information and to make quick and accurate decision. Processing speed is more than a simple reaction to

(16)

visual stimuli, because visual motor coordination and sustained focus is required to perform well in this area. Even though Mayes and Calhoun (2003) were able to identify children with high-functioning ASD that had obtained weak results in FDI and PSI Indexes, with 73 percent accuracy, this consistently found profile is not used as a diagnostic tool. Still, it is clear that the core distinctive index from individuals with ASD or without ASD is the PSI. Indeed, impairments in planning ability and visual motor coordination are usually present in children with ASD. Several studies suggest that low PSI scores only result from visual motor coordination impairments and motor speed difficulties, but not from ‘cognitive’ processing speed difficulties. Scheuffgen, Happé, Anderson, & Frith (2000) found that children with ASD demonstrated surprisingly fast processing speed abilities in inspection time tasks, a processing speed measure free from motor demands. Also in the study of Wallace, Anderson, & Happé (2009) children with ASD matched typically developing children on inspection time task results.

The processing speed is further related to working memory: increased processing speed can decrease the load placed on working memory, while decreased processing speed can impair the effectiveness of working memory (Wechsler, 2003a). In the study of Rosa et al. (2017), children with ASD showed lower verbal working memory and non-verbal working memory results than their siblings without ASD. Schaeffer (2018) indicated that the performance on verbal and non-verbal working memory strongly differs in children with ASD, in that they only show a weaker performance on verbal memory tasks. For verbal memory, studies have shown that individuals with ASD have intact semantic memory but that they have problems in memory for more complex linguistic material. A recent study of Williams, Minshew, Goldstein and Mazefsky (2017) gave evidence for long term memory problems for stories in adults with ASD. Those participants with ASD were poorer at recall of story details and thematic information in comparison with typical developmental adults. It seemed that the participants with ASD had difficulty with the use of organizational strategies to remember the stories.

Executive functioning, Processing speed, including visual motor skills and verbal memory (verbal working memory and verbal long term memory) may have important implications on the neuropsychological functioning of children with ASD. The question is if there is a direct association between the neuropsychological indexes ‘processing speed,’ ‘working memory’, ‘long term memory’ and social functioning. In the study of Oliveras - Rentas et al. (2012)

(17)

correlations are reported between processing speed difficulties and communication difficulties in children with ASD. In contrast, Black, Wallace, Sokoloff, & Kenworthy (2009) found that social skills are not directly associated with the results of the non-verbal indexes, but rather with the discrepancy score between the verbal and nonverbal index score. Moreover, in another study of Malhi and Singhi (2015), only the global IQ was positively correlated with adaptive functioning in the social domain. In fact, the global IQ emerged as the main predictor of social adaptive functioning. At any rate, the association of particular cognitive abilities with the core symptomatology of ASD, namely the social deficits, remains unclear. Hence, in this study not only IQ measurements are investigated, but a broad spectrum of neuropsychological functions are taken into account. Neuropsychological measurements with different cognitive tasks may affect the ‘intellectual’ profile characteristic of ASD and may reveal different associations with social symptomatology.

2.4 Social Functioning in Children with Developmental Language Disorder

Another question is then if the co-occurring social interaction problems in children with DLD are related to different cognitive deficits. Before answering this question, it is important to understand the relationship between the development of language and the development of social communication and to what extent a language impairment cause deficits in social functioning.

Poor social skills, including poor pragmatic skills represent a major area of concern for children with DLD (Botting, & Conti-Ramsden, 2008). Children with DLD may display poor conversational turn-taking, low rates of verbal communication with peers, high rates of ignoring of peers, and decreased interactive play (McCabe & Marshall, 2006).

2.4.1 Three Different Models of Social Communication of Bishop

Bishop (1997) discussed three different models that could account for the link between language impairment and poor social communication and interaction.

The first of these models suggests that the deficits seen in communication in DLD are the result of general and nonspecific limitations in working memory and phonological processing capacity. Working memory impairments are commonly reported for children with DLD, including poor performance on phonological storage tasks, requiring immediate repetition

(18)

(Archibald & Gathercole, 2007; Baddeley, 2003; Coady & Evans, 2008). Disproportionate deficits on complex span tasks, in the study of Archibald and Griebling (2016), were seen as evidence for additional impairments in the processing component of the working memory. Griffiths (2007) also reported a correlation between pragmatic skills and phonological short term memory and working memory scores. One might hypothesize that reduced abilities in automatized language processing might cause overload in the working memory and result in misunderstanding and difficulties at the pragmatic level.

In the second model discussed by Bishop (1997) problems with social interaction seen in children with DLD are considered as a consequence of their distorted social experiences, resulting from their difficulties with language and communication. Evidence of the adverse effect of language disorders upon social interaction has been described for children with DLD (Brinton & Fujiki, 2004). Their repertoire of socioemotional behaviors reflects social adaptations to their language limitations. Specifically, the social interaction problems of children with DLD are related to poor collaborative skills (e.g., negotiation), difficulties in establishing and maintaining friendships, reduced conflict resolution, and even showing withdrawal behaviors on the playground or peer victimization (Gibson et al., 2013). Recent longitudinal studies show that the long-term social problems of these children can increase during their transition into adolescence (Conti-Ramsden, Mok, Pickles, & Durkin, 2013) and that through a kind of bidirectional model, peer rejection may limit the acquisition of social skills (Banerjee, Watling, & Caputi, 2011).

The third model of Bishop (1997) includes a more alternative view: that there is some underlying deficit in social cognition, that creates difficulties in social interaction. Social cognition allows people to understand and report one’s own mental states and those of others, and it is essential to explain and predict people’s behavior (Hughes & Leekam, 2004). Certain aspects of language development are important for configuring social cognition, such as semantics (e.g., understanding mental verbs like ‘believe’ or communicational verbs like ‘say’), syntax and grammar (e.g., comprehension of sentential complements embedded under a mental or communicational main verb), or pragmatics (e.g., being successful at connected communication) (Milligan, Astington, & Dack, 2007). As children with DLD have difficulties in these structural language abilities, a delay in social cognition has been seen in this group. In fact, children with DLD showed therefore a poorer performance on false belief tasks (Andrés-Roqueta, Adrian, Clemente, & Katsos, 2013).

(19)

2.4.2 The Procedural Deficit Hypothesis

Another recent, new hypothesis is that implicit learning impairments may contribute to the social-communicative and behavioral atypicalities associated with DLD, by making it more difficult for individuals with DLD to extract patterns from the environment, in order to understand the unspoken rules governing language and social mores (Klinger, Klinger, & Poligh, 2007). Ullman (2004) called it the procedural deficit hypothesis, in which he claims that impairments in rule-based processes in language development and in social interaction may be explained by deficits in the neural networks that underpin procedural memory. These deficits contain neurological abnormalities affecting the frontal/basal ganglia and cerebellar circuits that are important for the procedural memory system and are observed across a range of developmental disorders, including DLD. In contrast, impaired implicit learning (also called statistical learning) fails to account for the social-pragmatic difficulties associated with ASD (Obeid et al., 2016).

Anyhow, there is still no consensus about the risk factors underlying the social impairment seen in children with DLD. Briefly, the problems in social interaction among children with DLD could be due to deficits in structural language affecting reducing capacities in working memory and phonological short term memory, affecting distorted social experiences and social cognition or to implicit learning.

2.5 Neuropsychological Profile of Children with Developmental Language Disorder

In recent years DLD has been approached from a neuropsychological perspective, that is, DLD has been associated with neuropsychological difficulties (Rapin, Dunn, & Allen, 2003).

Due to delayed language skills, children with DLD obtain lower scores on verbal tasks. Therefore, a discrepancy between the verbal intelligence quotient (VIQ) and performance intelligence quotient (PIQ) is common in children with DLD (Finnish Association of Phoniatrics and Finnish Association of pediatric Neurology, 2010). However, in the DSM-5 criteria (APA, 2013) a PIQ within the normal age range is no longer included. This is because there is controversy around the non-verbal IQ development in children with DLD. The explanations given to the difference regarding non-verbal IQ vary in authors such as Newton (2010) and Spaulding, 2010 (Gallinat & Spaulding, 2014). Spaulding (2010) state the

(20)

possibility of the existence of non-linguistic deficits associated in children with DLD. In particular, in two meta-analyses, children with DLD obtained lower PIQ scores than age-matched typically developing children (Earle, Gallinat, Grela, Lehto, & Spaulding, 2017; Gallinat & Spaulding, 2014). Lum and Bleses (2012) and Gallinat and Spaulding (2014) pose the existence of common mechanism in the verbal and non-verbal IQ development, such as impairments in the procedural memory, as has also been discussed above in the procedural deficit hypothesis suggested by Ulmann (2014). In the recent study of Schaeffer (2018) a relationship between weak nonverbal performance scores and problems in the grammar component of language in DLD is described. In this respect, the children with DLD clearly distinguish themselves from the children with ASD, who show no weaknesses in either grammar or nonverbal working memory.

However, it seems there is no specific dysfunction for DLD, but rather various cognitive skills that may present deficits (Buiza-Navarette, Adrian-Torres, & Gonzalez-Sanchez, 2007). In addition, there is increasing evidence that DLD is associated with executive dysfunctions such as impaired inhibition and weakened fluency, updating, planning, and working memory (Bishop & Norbury, 2005; Henry, Messer & Nash, 2011). Above all, impairment of verbal working memory and phonological short-term memory on tasks like sentence and nonword repetition, are considered the clearest risk markers of DLD (Archibald & Gathercole, 2006). Therefore, auditory processing (including PSTM) and working memory indexes are expected to be low in children with DLD. Indeed, the study by Meir and Armon-Lotem (2014) shows that peers with a typical development outperform children with DLD, on the forward digit span, nonword repetition, and a sentence repetition tasks. Additionally, they report associations between language proficiency and verbal working memory. This is not all that surprising, since verbal working memory contains a language component, and if language is weak, this may affect the performance on a working memory task including language.

Surprisingly neither the performance intelligence, nor other cognitive functions (nonverbal reasoning, nonverbal inhibition, theory of mind), nor grammar impairments could explain the social deficits in children with DLD (Schaeffer, 2018). For the reasons mentioned before, the present study argues that especially overall language performance problems, which could be cause by a weak auditory processing (PSTM) or verbal working memory, are related to social functioning in the group with DLD.

(21)

2.6 The Present Study

In sum, both groups ASD and DLD have overlapping social-communicative difficulties. Bishop (2010) assumed therefore that that ASD and DLD are different conditions of the same spectrum. The current study focuses on this question: Should DLD and ASD be distinguished as different impairments, or do they show so much overlap that they cannot really be separated and should be considered as instantiations on the same continuum? In this study a large and diverse test battery is used to reveal the neuropsychological profiles of both groups. If the neuropsychological profile of the children with ASD strongly differs from those of children with DLD, than this suggest that both neurodevelopmental disorders are not from the same continuum. After investigating the neuropsychological functions of both groups, we investigated if the social impairments of children with ASD can then also be attributed to a different cause than the social deficits seen in DLD. The hypothesis is that children with ASD have different strength and weakness in their neuropsychological functioning than children with DLD. Next we hypothesize that commonalities in social profiles across ASD and DLD can be attributed to different underlying neuropsychological weaknesses.

The present study compared the indexes and subtest scores of a broad neuropsychological investigation according to the Cattell–Horn–Carroll model (commonly abbreviated to CHC-model) between children with ASD and DLD. The CHC-model is widely accepted as the most comprehensive and empirically supported investigation of neuropsychological abilities (Kaufmann, 2009, p. 91). With a cross-battery approach, including subtests of the Wechsler Intelligence Scale for Children (WISC-III; Wechsler, 2005), the Clinical Evaluation of Language Fundamentals-4-NL (CELF; Kort, Schittekatte & Compaan, 2008), the Snijders-Oomen Nonverbal Intelligence Test (SON-R 2.5-7, Tellegen, Winkel, & Laros), the NEPSY-II (Korkman, Kirk, & Kemp, 2007a) and the Children Memory Scale (CMS; Cohen, 1997) the following broad abilities are tested in both children with ASD and children with DLD: Comprehension, Fluid Reasoning, Working Memory, Long-Term Memory, Visual and Auditory Processing, and Processing Speed. In this model, the above-named broad abilities scores are calculated with scores obtained by two or three subtests. In total, there are 15 subtests, more precisely: analogies, categories, information, vocabulary, similarities, digit span, recalling sentences, block design, picture completion, nonsense words, logical memory, word associations, logical memory delayed response, coding, and symbol search.

(22)

The expectations for the neuropsychological results in the group with ASD are the following: - We expect that children with ASD’s working memory is relatively weak in

comparison with their global intellectual functioning, as was shown in the study of Rosa et al. (2017).

- Children with ASD obtained higher scores on the Verbal Comprehension Index (VCI) and the Perceptual Observation Index in comparison with the Processing Speed Index (PSI) and Freedom From Distractibility Index (FDI) in the study of Mouga et al. (2016). We expect similar results in this study. More precisely we expect higher scores on both Comprehension and Visual Processing than on Processing Speed and Working Memory in the group with ASD.

- We expect weak scores on Long Term Memory, as the study of Williams et al. (2017) showed problems in long term memory of stories in adults with ASD.

The expectations for the neuropsychological results in the group with DLD are the following: - Our expectations for the children with DLD are based on the first model of Bishop (1997) and on the study of Archibald and Gathercole (2016), in which working memory and auditory processing are weak in children with DLD. We expect to replicate these results by observing weak scores on both components Working Memory and Auditory Processing in relation to the other neuropsychological components.

- We expect that the group with DLD achieves higher scores on non-verbal components, in particular on Visual Processing and Fluent Reasoning in relation to indexes that require language skills (e.g. Comprehension). These expectations are in line with the commonly reported discrepancy between the verbal intelligence quotient (VIQ) and performance intelligence quotient (PIQ) in children with DLD (Finnish Association of Phoniatrics and Finnish Association of pediatric Neurology, 2010).

The next aim of the present study is to investigate the relationship between social interaction abilities and the neuropsychological abilities, to see if social impairments are related to different underlying cognitive deficits in both groups. The social interaction is evaluated by the Social Responsiveness Scale (SRS; Constantino & Todd, 2005), a parent report questionnaire to assess the degree of the social impairment. It includes items that ascertain social awareness, social cognition, social communication, social motivation, and autistic mannerisms. Correlations are calculated between these domains of the questionnaire and the

(23)

index scores of the neuropsychological assessment (CHC-model). As the gold standard assessment for ASD, namely the Autism Diagnostic Observation Scale (ADOS-2; Lord et al., 2012) and the Autism Diagnostic Interview-Revised (ADI-R; Rutter et al., 2003) was also taken for the diagnostic procedure of the children with ASD, the social scale ‘reciprocal interactions’ of these tests will also be included in this study, but only for the group with ASD.

The expectations for social functioning and its relationship with the neuropsychological functioning in the group with ASD are the following:

- We expect that reduced opportunities to learn social interaction skills in the group with ASD is due to impairment in processing speed and working memory, which has also been described in the studies of Ament et al. (2015), Chien et al. (2015) and Oliveras-Rentas et al. (2012). Higher scores on the SRS or ADI-R mean a stronger severity of impairments in the social domain. This means that a negative correlation is expected between the scores of the SRS, the ‘reciprocal interactions’ scale of the ADI-R, and both Working Memory and Processing Speed components in the group with ASD.

- We expect that children with a higher global intelligence have fewer problems in interpreting social situations. In other words, we expect a negative correlation between global intelligence and social cognition on the SRS, as has also been seen in the study of Malhi and Singhi (2015).

The expectations for social functioning and its relationship with neuropsychological functioning in the group with DLD are the following:

- We expect that deficits in structural language and comprehension of language have a negative influence on social skills, like has been shown in the study of Milligan et al. (2017). Therefore negative correlations between the comprehension index, that requires acquired knowledge of language, and social communication scores are expected.

- We expect negative correlations between the SRS scores and both auditory processing and working memory. This agrees with Bishop (1997), who stated in her first model that the reduced opportunities to learn social interactions could be due to impairment in phonological processing and working memory.

(24)

In sum, the possible overlap in the social profile of ASD and DLD and the suggestion that their common social impairments come from different causes will be explored by:

(1) Identifying whether children with ASD carry different cognitive characteristics than children with DLD

(25)

3. Method

3.1 Procedure

Data were collected from a database on archival research of a multidisciplinary center for children with neurodevelopmental disorders in Flanders, the Dutch speaking part of Belgium. To be included in this study, all participants had to be given a broad neuropsychological assessment (according to the CHC-model) and the participants primary caregiver had completed the Social Responsiveness Scale (Constantino & Todd, 2005) survey form.

The neuropsychological test data and SRS interview scores included in this study come from children that were assessed and diagnosed with ASD or DLD by a multidisciplinary team between the years 2015 and 2019. A total of 101 records of school-aged children, ranging from 6 to 16 years and 11 months, met the inclusion criteria and were included in this study. Participants were divided in two clinical groups: children with ASD (n = 75; mean age = 9.8 year; 57 male/18 female) and children with DLD (n = 26; mean age= 8.5 year; 16 male/ 10 female).

Psychiatric classifications were made according to DSM-5 criteria, by a multidisciplinary team including an experienced neuropsychiatrist and psychologists, on the basis of clinical structured and semi-structured interviews (with the caregivers), observations of the child, questionnaires, and neuropsychological assessments.

3.2 Participants

3.2.1 Children with ASD

The ASD diagnosis was assigned on the basis of the gold standard instruments: parental or caregiver interview taken by the Autism Diagnostic Interview –Revised, ADI-R (Lord et al. 1994), and the direct structured assessment, the Autism Diagnostic Observation Schedule, ADOS (Lord et al. 1989). All ASD patients had positive results in the ADI-R and ADOS for ASD, and met the criteria for ASD from the DSM-5. The exclusion criteria for the children with ASD, who participated this study were evaluated through the anamnesis carried out with the families. They included neurological or genetic diseases, brain lesions, sensory, auditory or motor deficits and IQ below 80. Further children with comorbidities with the neurodevelopmental disorders DLD and ADHD were excluded, as the typical, distinct

(26)

neuropsychological profiles in these neurodevelopmental disorders could have influenced the results of present study.

3.2.2 Children with DLD

All children in the group of DLD met the criteria of a language disorder from the DSM-5. In addition all participants spoke Dutch as their primary language, passed a hearing screening performed at 20dB and achieved average standard scores (>80) on performance IQ. Use of nonverbal intelligence in the diagnosis of DLD is debated, but remains a commonly used diagnostic criterion for DLD and is important in this study as we compare children with DLD with children with high functioning ASD. Language abilities were assessed for all participants via core language subtests on the Clinical Evaluation of language Fundamentals, Fourth Edition (CELF-4-NL) (Kort, Schittekatte & Compaan, 2010). All participants achieved standard scores at least 1.25 SD (standard deviations) belows the mean of age-matched children’s language scores on the Core Language Score (CLS) or on at least two sub index scores or showed a severe deficit (e.g. 2SD or more below the mean) on any language measure. The Core Language Score is based on performance on four core subtests form the CELF-4 including Concepts and Following Directions, Recalling Sentences, Formulated Sentences, and depending on the age of the child, Word Structure (under 9 years) or Word Classes (9 years or older). The exclusion criteria included hearing or visual impairment, neural-motor impairment and psychiatric or behavior disorder other than deficits in learning. Children with a comorbid ADHD or ASD were excluded from the study.

3.3 Materials

In addition to measures used as inclusion criteria, the dataset of the participants had to contain data of the neuropsychological assessment (CHC-model) and the Social Responsiveness Scale (Constantino & Todd, 2005) survey form. Both assessments are discussed below, in addition to the golden standard measures for ASD diagnosis, namely ADOS and ADI-R.

3.3.1 Neuropsychological assessment and the Catell Horn-Carroll (CHC) cognitive abilities model

The following is a brief overview to explain which tests are used and how the results are interpreted in the context of the CHC-model. The CHC-model comprises a cross-battery

(27)

assessment, in which 15 different subtests are used. Index scores relating to 7 different specific cognitive areas can be interpreted.

1. The Fluid Reasoning index consists of two subtests, that examine abstract reasoning, logical thinking skills and problem solving abilities, which do not require verbal language production and comprehension:

- Categories (SON-R 2.5-7, Tellegen, Winkel, & Laros)

In this subtest the subject is shown three drawings of objects or situations that have something in common and has to discover the concept

- Analogies (SON-R 2.5-7, Tellegen, Winkel, & Laros)

This subtest consist of geometrical figures with the problem format A : B = C : D. The subject is required to discover the principle behind the transformation A : B and apply it to figure C.

2 Comprehension is typically described as a person's breadth and depth of acquired knowledge of the language, information and concepts of a specific culture, and the application of this knowledge. Following three subtests test comprehension knowledge; - Information (WISC-III-NL; Wechsler 2005)

In the first subtest answers have to be given to a broad range of general-knowledge topics (e.g. who was Einstein?)

- Vocabulary (WISC-III-NL; Wechsler 2005)

The participant is required to define orally presented words in this subtest. In this test word knowledge and verbal concept information are measured.

- Similarities (WISC-III-NL; Wechsler 2005)

In this subtest the participants have to describe how two given concepts are alike. The participants may for example be asked how a bicycle and a car are alike. This test is designed to assess verbal reasoning and the development of concepts.

(28)

3 Short-term memory includes memory spans and working memory of children. It is tested by two memory span tasks:

- Digit Span (CELF-4-NL; Kort, Schittekatte & Compaan, 2010)

The subtest ‘Digit Span’ asks for the ability to perform relatively simple manipulations, in particular to name the digits backwards, and impose cognitive demands on the working memory capacity.

- Recalling Sentences (CELF-4-NL; Kort, Schittekatte & Compaan, 2010)

The participants have to listen to spoken sentences of increasing length and complexity, and have to repeat the sentences without changing the sentence structure, word meaning, word structure, context meaning. In other words, the participants have to encode verbal information, maintain it in primary memory and immediately reproduce information.

4 Auditory processing circumscribes a wide range of abilities involved in the interpretation

and organization of phonological or verbal information:

- Nonsense Words (NEPSY-II; Korkman, Kirk, & Kemp, 2007a)

In this subtest, the participants have to repeat nonsense words, with increasing complexity, aloud. Discriminating patterns in sounds and the ability to store and recall nonsense-words is tested by this subtest.

- Logical Memory (CMS; Cohen, 1997)

Logical memory tests the ability to attend, register and immediately recall elements of a told story. This test consists of two short stories each containing 25 story units, which are verbally presented by the examiner. Immediately after each story, the participants are asked to repeat what they remember as accurately as possible.

5 Visual processing can be defined as the ability to make use of mental imagery to solve problems. It includes the ability to rotate, reverse and manipulate spatial configurations. - Block design (WISC-III-NL; Wechsler 2005)

(29)

The block design subtest is thought to tap spatial visualization ability and motor skill. The participants have to rearrange blocks that have various color patterns to match a pattern.

- Picture completion (WISC-III-NL; Wechsler 2005)

The participants are shown a picture in which there is a significant part missing, and are required to say what is missing. The ability to observe details and recognize specific features of the environment is measured by this task.

6 Long-term memory is the ability to store and consolidate new information in long-term memory and later fluently retrieve the stored information. It is tested by two subtests:

- Word associations (CELF-4-NL; Kort, Schittekatte & Compaan, 2010)

In this subtest the participants have to produce isolated words from a specific semantic category.

- Logical Memory Delayed Response (CMS; Cohen, 1997)

This subtest tests the ability to retain and recall information of a meaningful story or connected discourse. The participants are asked to recall the story (of the subtest Logical Memory) again after thirthy minutes delay (delayed recall).

7 Processing speed requires high mental efficiency (i.e., attention and focused concentration). It examines the ability to rapidly and accurately search, compare for visual similarities. Therefore the ability to quickly recognize simple visual patterns and the ability to scan, compare and look up for visual stimuli is important in this index. Further the tasks impose additional cognitive demands on the immediate short-term memory. This index is sensitive to motivation, to difficulty working under a time pressure and to motor coordination. The processing speed index score is based on the following subtests:

- Coding (WISC-III-NL; Wechsler 2005)

The participants are presented with a key in which numbers 1 to 9 are each connected with a different symbol. They have to use this key to put in the correct symbols for a list of numbers.

(30)

- Symbol search (WISC-III-NL; Wechsler 2005)

The participants have to look at two target symbols and then have to examine a group of symbols to see if the target symbols are repeated.

3.3.2 The Social Responsiveness scale (SRS; Constatino & Todd, 2005)

The SRS is a validated, widely used questionnaire that assesses behavioral and social-communicative traits. The SRS form contains 65 questions about a child’s behavior and is completed by a parent. The questions can be answered on a four-point Likert-scale. Answer possibilities range from 0 (never true) to 3 (almost always true). The higher a child scores in total, the more social problems or autistic traits it is believed to show. The SRS manual divided T-scores (M=50, SD =10) in four levels. (0) High level of social responsiveness (T-scores of below 40), (1) normal level of social responsiveness (T-score between 40 and 60), (2) mild to moderate deficits in social responsiveness (T-scores between 61 and 75) and (3) severe deficits in social responsiveness (T-scores above 75). When children score within group 0 or 1, so they score below 60, it suggests that their social responsiveness does not show any deficits. When a child scores within group 2 or 3 however, their social responsiveness does show deficits. There are different scoring ways for boys or girls. Five subscales are provided: Social Awareness, Social cognition, Social Communication, Social Motivation, and Autistic Mannerism. The subscale ‘Social Awareness’ represents the sensory aspects of reciprocal social behavior. The items in this subscale question the ability to pick up on social cues. The ability to interpret social cues is examined by the items of ‘Social Cogniton’. This category represents the cognitive- interpretive aspects of reciprocal social behavior. The subscale ‘Social Communication’ includes expressive social communication. The extent to which a respondent is motivated to engage in social- interpersonal behavior is included in ‘Social Motivation’. Finally the items of the subscale ‘Autistic Mannerisms’ include restricted interests and stereotypical behavior. The SRS items of each subscale are represented in Appendix I. The SRS total score and scores from each subdomain are used in this study.

3.3.3 Autism Diagnostic Observation Schedule, version 2 (ADOS-2; Lord et al., 2012)

The ADOS-2 (Lord et al., 2012) is a standardized diagnostic measure. This assessment is considered gold standard in research protocols and is the most commonly used standardized

(31)

diagnostic measure in both research protocols and clinical practice. It has strong psychometric properties, including high reliability and validity. The ADOS-2 examines behaviors in two distinct domains, Social Affect (SA) and Restricted and Repetitive Behaviors’ (RRB). The ADOS-2 total score (TS) is obtained by the sum of SA and RRB scores. The protocol consists of series of play-based structured and semi-structured tasks that involve social interaction.

3.3.4 Autism Diagnostic Interview-Revised (Dutch language version) (ADI-R; de Jonge, de Bildt, Le Couteur, Lord, & Rutter, 2014)

The ADI-R is a semi-structured interview conducted by a clinician to collect information from the child’s primary caregiver on aspects of a child’s behavior. It consists of 93 questions that are based on DSM-5 diagnostic criteria. The questions cover three main areas:

(1) Qualitative or reciprocal social interactions; (2) communication and language; and (3) restricted and repetitive stereotyped interests and behavior (Lord et al., 1993). In this study, only the items of the first two subdomains ((1) and (2)) are used. The items of these subdomains are presented in Appendix II.

3.4 Data analysis

Statistical analyses were performed by the version 24 for Microsoft Windows of the Statistical Package for Social Sciences software (SPSS; Chicago, IL, USA). The verification of the assumptions of normality for the application of parametric tests on the variables of interest was done by a Kolmogorov–Smirnov test with Lilliefors correction.

A large amount of missing data points in the neuropsychological measurements prevented us from performing a repeated measures ANOVA analysis. The default approach to missing data in repeated measures analysis is Listwise Deletion, which drops all observations with any missing data on any variable involved in the analysis. This would have led to a too small number of participants. In a mixed approach only the missing data points are dropped, while remaining data will be retained. Therefore differences within the neuropsychological profiles of children with ASD or DLD were tested using repeated measures mixed models with an unstructured covariance structure. The fixed factors ‘diagnose (ASD or DLD)’, ‘Neuropsychological components’ (Fluent Reasoning, Comprehension, Working Memory, Visual Processing, Auditory Processing, Long Term Memory, Processing Speed) and the

(32)

interaction between diagnose and neuropsychological components were used to perform this analysis. Next, differences between the SRS subdimensions were tested also using repeated measures mixed models with an unstructured covariance structure.

Additionally, Spearman correlations were calculated to determine if an association exists between the neuropsychological components and SRS subdimensions, in both groups. A correlation between each neuropsychological component and ASD symptomatology (Language/Communication, Reciprocal Social Interactions, and Repetitive Behaviors/Interests results from ADI-R and ADOS) was calculated only for the ASD group.

(33)

4. Results

4.1 Descriptive Statistics

Average scores and standard deviations of the neuropsychological components and subtests scores from the two clinical groups (ASD and DLD) are reported in Table 1.

Table 1

Average index’s and subtest scores of the neuropsychological abilities of the group with ASD and the group with DLD

Note. ASD = Autism Spectrum Disorder, DLD = Developmental Language Disorder, N = number of

participant, M = mean, SD = standard deviation, SON-R 2.5.7 = Snijders-Oomen Nonverbal Intelligence Test (Wechsler scales), WISC-III-NL = Wechsler Intelligence Scale for Children 3th edition (Wechsler scales) ,CELF-4-NL= Clinical Evaluation of Language Fundamentals-4-NL (Wechsler scales), CMS = Children Memory Scale (Wechsler scales).

A Kolmogorov–Smirnov test with Lilliefors correction was done to verify of the assumptions of normality and indicated that each neuropsychological component follows a normal distribution; Fluent Reasoning D(90) = 0.09, p = 0.08, Comprehension D(90) = 0.07, p =

ASD DLD N M SD N M SD Fluent Reasoning 64 101 15.6 19 98 4.9 Analogies (SON-R 2.5-7) Categories (SON-R 2.5-7) 50 10 1.6 12 9 2.3 50 9 1.8 12 10 1.1 Comprehension Knowledge 63 106 17.2 21 93 7.6 Information (WISC-III-NL) Vocabulary (WISC-III-NL) Similarities (WISC-III-NL) 51 10 1.5 13 8 1.2 52 10 2.1 17 7 1.2 47 12 1.7 15 8 1.3 Working Memory 64 97 14.2 16 86 8.7

Digit Span (CELF-4-NL)

Recalling Sentences (CELF-4-NL)

51 9 2.2 7 7 1.2

49 9 1.4 12 7 1.3

Visual Processing 64 106 22.1 21 98 9.5

Block Design (WISC-III-NL)

Picture Completion (WISC-III-NL) 52 ₅₂ 11 ₁₁ 1.2 _0.9 19 ₁₇ 10 ₁₁ 1.1 _1.1

Auditory Processing 41 100 14.6 11 86 8.4

Nonsense Words (NEPSY-II)

Logical Memory (CMS) 46 ₄₆ 11 ₁₀ 1.5 _1.6 6 ₇ 7 ₈ 1.4 _1.3

Long Term Memory 43 100 18.0 10 89 5.4

Word Associations (CELF-4-NL)

Logical Memory Delayed Response (CMS)

48 11 2.1 9 8 1.3

46 8 1.5 12 6 1.4

Processing Speed 64 96 17.4 19 98 12.1

Coding (WISC-III-NL) Symbol Search (WISC-III-NL)

52 9 1.4 20 8 1.2

52 10 1.4 22 9 1.2

(34)

0.20, Working Memory D(85) = 0.08, p = 0.20, Visual Processing D(91) = 0.09, p = 0.08, Auditory Processing D(50) = 0.17, p = 0.18 Long term memory D(56) = 0.08, p = 0.20 and Processing Speed D(88) = 0.06, p = 0.20.

Scores of the Social Responsiveness Scale (Constatino & Todd, 2005) were available for 64 children with ASD and 15 children with DLD. Average T- scores (M) and standard deviations (SD) of both the total and the subdomain scores are presented in Table 2. For interpreting these T-scores (M=50, SD =10), four levels of social responsiveness are described; (1) High level of social responsiveness (T-scores of below 40), (2) normal level of social responsiveness (T-score between 40 and 60), (3) mild to moderate deficits in social responsiveness (T-scores between 61 and 75) and (4) severe deficits in social responsiveness (T-scores above 75).

Table 2

Average total T-scores and subdomain T- scores of the SRS in the group with ASD (N=64) and the group with DLD (N=15)

SRS subdomains ASD DLD M SD M SD Social Awareness 62 21.5 60 18.3 Social Cognition 74 15.6 73 19.2 Social communication 76 17.5 73 16.6 Social Motivation 73 15.4 66 18.8 Autistic Mannerisms 82 13.5 65 19.0 Total score 81 14.2 74 16.8

Note. ASD = Autism Spectrum Disorder, DLD = Developmental Language Disorder, N = number of participant,

M = mean, SD = standard deviation.

A Kolmogorov-Smirnov test with Lilliefors correction indicates that the SRS subdimensions follow a normal distribution; Social Awareness D(73) = 0.09, p = 0.06, Social Cognition D(73) = 0.10, p = 0.06, Social Communication D(73) = 0.12, p = 0.06, Social Motivation D(73) = 0.09, p = 0.08, Autistic Mannerisms D(73) = 0.100, p = 0.07 Total Score D(73) = 0.07, p = 0.20

In Table 3 average scores (M) and standard deviations (SD) of the ADI-R (de Jonge, de Bildt, Le Couteur, Lord, & Rutter, 2014) and the ADOS-2 (Lord et al., 2012) are presented for the group with ASD.

Social Competence and its Relationship with the Neuropsychological Profile in Children with Autism Spectrum Disorder and Children with Developmental Language Disorder