Autism spectrum disorder (ASD) is an early onset neurodevelopmental disorder characterized by deficits in social communication, social interaction and restricted/repetitive behaviors or interests, which seriously affect the patient's social, academic, family, and other aspects [1]. To date, the etiology of ASD is unclear. It is believed that the pathogenesis of ASD is a result of an interaction of genetics and environment [2–3]. A recent Chinese multi-center study estimated that the prevalence of ASD is as high as 0.7%, with a male to female ratio of about 3:1 [4]. At present, the diagnosis of ASD is mainly based on behavioral manifestations. Although there is great heterogeneity between and within individuals, the core characteristics of ASD can still be reliably identified [5]. Although in the United States, the median age of ASD diagnosis remains stubbornly at 4 to 5 years [6], early warning signs can be observed in infancy [7]. Longitudinal studies of individuals with ASD have identified high-risk behaviors such as poor eye contact, lack of visual tracking, failure to respond to names, poor imitation skills, lack of social interest, and limited language ability [8–9]. Studying the early behavioral signs of individuals with ASD is the first step toward early identification, referral, diagnosis, and intervention. A growing number of studies support the efficacy of early behaviorally based interventions in ASD for the improvement of social communication and behavioral functioning to varying degrees [10–12].
Pragmatic language and ASD‑related characteristics
Pragmatic language refers to the ability to use spoken language to effectively convey information in different social contexts [13], which is frequently demonstrated to be a key component of the difficulties exhibited by individuals with ASD [14–16] .The new DSM-5 diagnosis of Social (Pragmatic) Communication Disorder (SPCD) also aims to identify individuals with pragmatic disorders. Because pragmatic language depends on more sophisticated developments of speech and language among other factors, a formal diagnosis of SPCD is rarely made in children younger than 4 years of age, although difficulties and delays in social (pragmatic) communication can be assessed and identified prior to this age [17]. Indeed, early signs of ASD may involve early pragmatic communication such as non-typical use and understanding of gestures, and differences in attention to social peers and common topics. In addition, about 30% of the individuals with ASD can only obtain minimal language communication with age [18]. With respect to those who have developed spoken language, pragmatic difficulties often persist, characterized by difficulties in the attempt to communicate and narrow communicative behaviors [19]. Moreover, pragmatic language difficulties may persist into adolescence and adulthood in ASD, as evidenced by poor topic management during conversation, frequent appearance of off-topic information, unusual rhythm in speech, reduced social reciprocity and response to social cues from peers [20–21]. Studies have shown that the effects of pragmatic impairments are far-reaching, and found to be associated with a greater incidence of behavioral problems, especially destructive and externalizing behaviors [22]. Among children, the long-term negative effects of pragmatic impairments include poor peer relations [23], being excluded and rejected by peers in school [24–25], and having difficulties forming successful social relations in adulthood [26–27].
Miranda et al [28] examined the relationships between the social communication questionnaire (SCQ) domains (including reciprocal social interaction, communication, and restricted and repetitive behaviors (RRBs)) and pragmatic language controlling for sex and vocabulary in children with ASD, and showed that the SCQ reciprocal social interaction domain correlated significantly with the pragmatic index, whereas there was no significant relationship with RRBs. However, previous studies have examined the association between the degree and type of RRBs and levels of functioning in ASD, showing that children with higher levels of RRBs were more likely to present with less developed receptive and expressive language abilities, and lower adaptive communication and socialization skills [29].
Joint attention and pragmatic language
Evidence of brain development and key stages of neural plasticity in infancy lends strong support to the importance of early intervention [30]. Studies have shown that early intervention is efficacious in modifying cognitive and social-communicative outcomes and potentially improving developmental trajectories in toddlers with ASD [8–9, 11]. Longitudinal studies of ASD can help understand the association between early developmental trajectories and outcomes, and provide benchmarks for early interventions.
Joint attention (JA) is a tertiary interactive process composed of two social partners and an object, and is an important component in the development of early social communication in typically-developing infants [31]. JA is often defined by measures that assess a discrete set of children’s skills using structured tasks (i.e., whether an infant can use pointing and eye gaze to share attention to an object). As early as 6 months of age, infants’ gaze can follow the caregiver to a new location [32], and, by 12 months of age, they can establish JA in various ways, such as looking at the caregiver and the object alternately, reaching for the object, or lifting the object to show to others [33]. These behaviors can help them build shared concerns for an object or event with social partners [33]. Difficulties with JA occur in ASD from infancy, and they are often one of the earliest signs accompanying a diagnosis of ASD [34–35]. There are two types of JA: response to joint attention (RJA) and initiation of joint attention (IJA). RJA is considered a more involuntary process of attention following, which can be improved to a certain extent with the development in individuals with ASD, whereas IJA is considered a more active process to share an experience, which shows little change with age in individuals with ASD [36–37]. RJA and IJA are both critical for the development of language in early childhood, especially for ASD [38] When children engage in JA, they engage in a social situation that helps to associate words with meaning and develop interactive abilities [39].
Several prospective studies have investigated the correlation between early JA levels and later pragmatic abilities in children at risk for ASD. For example, in 2015, Gillespie-Lynch et al. (2015) [40] evaluated JA levels in high-risk children with ASD at 12 and 18 months and explored the relationship between early JA levels and children’s pragmatic and structural language abilities 6 years later. Their results showed that response to joint attention (RJA) at 18 months of age could predict structural language but not pragmatic language in high-risk children with ASD. Greenslade et al. (2019) [41] collected social and language data in children at risk for ASD from 14 to 36 months old, and evaluated their pragmatic levels using the Pragmatic Rating Scale-School-age (PRS-SA) adapted from the PRS [42] from 8 to 12 years old. They found that the JA levels at 14 months of age could predict later pragmatic abilities in children at risk for ASD at school age.
Joint engagement and pragmatic language
Previous studies have mainly focused on the importance of early JA for pragmatic language development, especially for siblings of children with ASD[40–41, 43]. However, neurotypical infants rarely look at the caregivers’ faces before attending to an object during social interactions in reality [44–45], and children with ASD may engage with others in ways that are not captured in standard measures of JA [46–47]. In contrast, rather than assessing discrete skills, joint engagement (JE) is evaluated with respect to mutual and extended caregiver-child interactions in which a child and social partner sustain active involvement in the context of toy play [48–49]. JE may be more practical for interventions than JA because JE focuses on the interaction between caregiver and child, whereas JA is more relevant to objects and events. Interventions can be designed around promoting JE between caregiver and child [50]. The importance of JE was confirmed by a recent study showing that JE was a major predictor of expressive language even when JA was considered[49]. Adamson distinguished between three types of JE: (1) Coordinated JE (CJE), the most developmentally advanced periods, which requires the infant to make eye contact with the social partner; (2) Supported JE (SJE), a state in which the infant is actively involved with the same object as the social partner without making eye contact; and, (3) Object Engagement (OE), in which the infant is exclusively engaged with objects and not with the social partner[51]. For typically developing (TD) children, SJE accounted for a large proportion of play during the first two years of life, and by the age of 12 months, CJE began to appear, with an increase at the age of 24 months [48]. Children with ASD aged 2 to 3 years had a lower proportion of CJE and spent more time involved in OE than age and speech level matched TD children [52]. However, children with ASD had a higher proportion of SJE in parent-child play interactions, compared to their TD peers [49]. Studies have shown that the language outcome of children with ASD is related to the frequency of SJE, but not closely related to CJE. One possible explanation might be that the frequency of CJE is lower than that of SJE, and it almost does not occur in the communicative interactions between children with ASD and their caregivers [49].
Structural language and pragmatic language
The form (such as phonology, vocabulary, and syntax) and content (semantics) of language represent structural language. Pragmatic language needs to be expressed through structural language and social context, and structural language is a key element of pragmatic language. Pragmatic language is closely related to structural language, supported by studies that have shown that deficits in structural language are common in children with ASD and are related to pragmatic impairment [53–55]. In addition, studies have shown that structural language is a key predictor of non-literal pragmatic forms of language use such as idioms and metaphors [56–57]. Also, a meta-analysis showed that when the ASD group was matched with the TD group in terms of structural language performance, no significant difference was found in performance on pragmatic tasks [58]. To summarize, these studies suggest the importance of structural language to pragmatic language.
Current study
In neurotypical development, early JA, broader social-communication behaviors (e.g., shared enjoyment, gesture use), and structural language are the theoretical precursors of advanced social cognition. Social cognitive skills, as well as the ability to combine appropriate vocabulary and nonverbal communication behaviors in efficient well-formed expressive messages, are critical to later effective pragmatic communication [59–60]. Although the effects of the above factors on pragmatic outcomes have been studied separately, they have not been jointly studied as predictors of pragmatic language in the same model in children with ASD. Thus, the relative contribution of early predictors of pragmatic development - JA and especially JE and/or structural language - in ASD is unclear. Therefore, the current study examined pragmatic communication functioning in preschool-age children with ASD and TD controls, as well as the association with autistic symptoms and the early predictors of this functioning in the ASD sample. The following research hypotheses were addressed. First, between-group differences in pragmatic communication functioning at preschool years were examined, comparing the ASD group and TD controls. More specifically, TD controls were hypothesized to demonstrate significantly better preschool-age pragmatic communication functioning than the ASD group. Second, correlations between preschool-age pragmatic communication and autistic symptoms were expected to be significant when conducted with the ASD sample. Finally, in regression models, the combined set of 24-month frequency of joint engagement and structural language abilities were expected to explain a significant amount of the variance in later pragmatic communication in the ASD sample.