Children with autism spectrum disorders (ASD) range from those who have no verbal language ability to those who can converse with others with varying degrees competence. However, regardless of their speech and language abilities they cannot properly use their knowledge for the purpose of relationships. Even though they may have better vocabulary and grammar knowledge than their peers, they have difficulty in reciprocal conversation (Bedford et al., 2013).

Numerous studies have found that children with ASD have relatively higher ability in expressive language than receptive language, as opposed to typically developing children. The receptive language of typical children advances their expressive language (Kover, McDuffie, Hagerman, & Abbeduto, 2013). Generally, people can understand more than they can speak.

There have been consistent research results about the atypical characteristics in receptive and expressive language ability in children with ASD. Weismer et al. (2011) found that toddlers with autism from ages 24 to 36 months have significantly higher age-equivalent scores in expressive language than receptive language. In contrast, Volden et al. (2011) found that a comparison group of toddlers with developmental delay without ASD displayed the opposite trend. Also, they confirmed the strength of expressive language over receptive language in a large sample of 2- to 4-year-old children with ASD. Seol et al. (in press) reported significantly lower scores in receptive language in ASD children compared to developmental language delayed (DLD) children aged between 20 months to 50 months. According to the researchers, the ASD group was an expressive-language dominant group, compared to the DLD group, which was a receptive-language dominant group. A relative weakness in receptive language abilities is also prevalent in preschool and school-age children with ASD. Loucas et al. (2008) have compared receptive and expressive language abilities between school-aged children with specific language impairment (SLI) and ASD. They found that receptive language was less impaired than expressive language in children with SLI, whereas children with ASD lacked a receptive language advantage. Hudry et al. (2010) found that preschool children with ASD have notably lower scores in receptive vocabulary. Charman, Drew, Baird, & Baird (2003) observed that a sample of 134 preschool children with ASD had an overall delay in vocabulary, with receptive vocabulary being impaired to a greater extent than expressive vocabulary.

Many reports have indicated an atypical relationship between language comprehension and production in children with ASD. However, the extent to which this relationship extends to all individuals with ASD is unclear. Some of these inconsistent results are reported in school-age children with ASD and in their vocabulary ability. Kjelgaard & Tager-Flusberg (2001) examined the language ability of 89 children with autism, of which 44 performed the Clinical Evaluation of Language Fundamentals (CELF; Semel, Wiig, & Secord, 1995) and a vocabulary test. These 44 children got higher scores than control group in the expressive language scores of the CELF but not the vocabulary test. It has been argued that children with ASD are more impaired in receptive language ability (Hudry et al., 2010), but one study showed that approximately 30% of the children with ASD had an atypical pattern of both receptive and expressive vocabulary on the MacArthur Communicative Development Inventories (MCDI; Fenson et al., 1993).

Then, how can we explain the relative weakness of children with ASD in receptive language? One possible theory can come from the mutual exclusivity (ME) principle in typical development of vocabulary acquisition. The ME principle refers to the assumption that novel words refer to unfamiliar objects or objects for which the child does not yet have a label (Bedford et al., 2013). Most parents and caregivers give ostensive and referential cues to young language learners with repetitive labeling of the name of the object. Word learning is subsequently demonstrated in the absence of such cues, in which case infants make use of various heuristics to infer a speaker’s referent, which is ME, and most children start to use this principle toward the end of their second year of life (Halberda, 2003). With or without ostensive and referential cues, whenever labeling happens and a child can associate an object with a word just heard, this kind of word learning context is called fast mapping. In addition, word retention tests following fast mapping show that more context helps children remember newly acquired words; this reinforcement of from the interaction partner, is called feedback (Horst & Samuelson, 2008).

Children with ASD are less responsive to social cues, in particular to referential cues, and some experimental studies have shown that they have difficulties in using cues for word learning. Bedford et al. (2013) studied the ability to use ME and retention of new words in 2-year-old children at risk of ASD. They found that the ME and fast mapping of children with ASD was intact (de Mrchena, Eigsti, Worek, Ono, & Snedeker, 2011) but that they showed differences when using feedback from interaction partners. Language delay in ASD children, especially in vocabulary, could be caused by their lower sensitivity to social-communicative cues; they do not benefit from ostensive feedback and therefore are less able to retain the results of word-object mapping (Baron-Cohen, Baldwin, & Crowson, 1997).

Language is also one of the most impaired areas of children with intellectual disabilities (ID). Studies indicate that children with ID often exhibit poorer language skills compared with typically developing (TD) children of the same mental age (Owens, 2010). The overall sequence of their development is similar to that of TD children before they use words, but their rate of development is slower. They develop intentions, role taking, and presupposition. Chapman, Bird, & Schwartz (1990) studied novel word learning in children with Down syndrome and chromosomal disorders and found out they are as skilled as their mental-age-matched TD peers in inferring novel word meanings and in producing these words correctly thereafter.

There have been mostly consistent research results about the characteristics of vocabulary of children with ASD and ID. Receptive vocabulary is greater than productive (i.e., expressive) vocabulary for typically developing children and conversely (that is, productive > receptive vocabulary) for atypically developing children (Hudray et al., 2010). Expressive vocabulary is less impaired than receptive in ASD children, but there is little research regarding receptive and expressive vocabulary for children with ID. Moreover, many research studies have been done using vocabulary tests from parents’ reports, which mostly compared ASD children with typically developing or language impaired (LI) children. In addition to comparison with typical and LI children, it is necessary to compare the atypical relationship in vocabulary among children with high and low functioning ASD, and ID children in clinical practice. The aim of this study is to find (1) the differences in expressive and receptive vocabulary abilities among children with high and low functioning ASD and ID children and (2) the differences between expressive and receptive vocabulary among three groups.

METHODS

Participants

The participants were 33 children between the ages of 60 months and 94 months who visited a child and adolescent psychiatry clinic. They were divided into three groups: high functioning ASD (HF ASD), low functioning ASD (LF ASD), and ID. Children with ASD and ID were diagnosed by a child and adolescent psychiatrist, using the criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text-revised (DSM-IV-TR). There are no available diagnostic guidelines for HF ASD and LF ASD. These are the most commonly used clinical descriptors to differentiate individuals with autism. Conventionally, if a child is diagnosed with ASD meeting the conditions of DSM-IV or -5, but displays cognitive ability in the low average to above average range, he or she is said to have HF ASD (Sansosti, Powell-Smith, & Cowan, 2010). In this study, low functioning autism children were diagnosed by the results of cognitive evaluation using the Korean Wechsler Preschool & Primary Scale of Intelligence (K-WIPPSI) and Korean Educational Development Institute Wechsler Intelligence Scale for Children (KEDI-WISC). Data from 55 children (16 children with HF ASD, 22 children with LF ASD, and 17 children with ID) was collected in the beginning of the present study, however, subsequently some children were excluded when they did not meet the criteria of Intelligence Quotients (IQ)—IQ under 40 and borderline IQ between 70 to 80. Vocabulary test scores are meaningless when IQ is too low and standard deviations are too large when borderline intellectual functioning children are included. After these exclusions, there were 10 children in the HF ASD group (9 boys and 1 girl), 11 children in the LF ASD group (8 boys and 3 girls), and 12 children in the ID group (8 boys and 4 girls). Group means and standard deviations for the chronological age of the children and their Verbal Intelligence Quotient (VIQ) and Performance Intelligence Quotient (PIQ) can be seen in Table 1. Mean ages were 76.70 months (range, 63-94 months) for the HF ASD group, 70 months (ranged, 61-87 months) for the LF ASD group, and 73.33 months (range, 60-90 months) for the ID group. Children with HF ASD had a 97.90 mean VIQ score (range, 82-139) and 102.50 mean PIQ score (range 90-134). The LF ASD group had a 67.18 mean VIQ score (range, 53-82) and 60.73 mean PIQ score (range, 49-66). The mean VIQ and PIQ scores of children with ID were 69.17 (range, 53-84) and 65.92 (range, 52-88), respectively. One-way repeated ANOVA was run to compare these characteristics across the three groups. Group comparisons revealed the following: (1) there was no significant difference in chronological age among the three groups and (2) both VIQ and PIQ in the HF ASD group were significantly higher than in the LF ASD and ID group (VIQ: F_(2,30)=18.00, p<.001; PIQ: F_(2,30)=45.65, p<.001).

Table 1.

Participants by group, age, VIQ, and PIQ

Language assessment

The receptive and expressive vocabulary test (REVT; Kim, Hong, Kim, Jang, & Lee, 2009), which is standardized Korean vocabulary test, was used in this study. The test consists of a receptive vocabulary test (REVT-R) and an expressive vocabulary test (REVT-E). The item sets are arranged in order of increasing difficulty based on Item Response Theory (IRT). In REVT-R, children are requested to point the appropriate picture among 4 choices; children are requested to name a picture as an examiner shows in REVT-E. The test is suitable for people ranging in age from 2 years old to over 16 years old. In the present study, data was comparing receptive and expressive vocabulary scores from age-equivalent children from other research studies (Abbeduto et al., 2001; Chapman, 2006; Krantz & McClannahan, 1993; Miyamoto, Houston, Kirk, Perdew, & Svirsky, 2003).

RESULTS

Group differences on scores of receptive and expressive vocabulary test

One-way repeated measures ANOVA was performed to compare age-equivalent scores on both the receptive and expressive vocabulary test (REVT) for the three participant groups. The between-subjects effect indicates that group differences were significant (F_(1,30)=384.55, p<.001), but the within-subjects effect and interaction effect (REVT by Group) were not statistically different (F_(1,30)=231.43, p>.05 and F_(2,30)=111.29, p>.05; respectively).

Additionally, MANOVA was used to compare groups with age-equivalent REVT scores and the LSD post-hoc test was conducted to make pairwise comparisons between two of the three groups with REVT scores. As can be seen in Table 2, performances on the receptive vocabulary of REVT was significantly different among the three groups (HF ASD, LF ASD, and ID) in terms of age-equivalent scores (F=13.68, p<.001). LSD post-hoc comparisons revealed significant difference between the HF ASD and the both LF ASD (p<.001) and ID groups (p=.002), but only slightly different between the HF ASD, LF ASD, and ID groups (p=.066). Also, age-equivalent scores (F=11.01, p<.001) on the expressive vocabulary of REVT were significantly different among the three groups. Posthoc tests showed children with HF ASD had consistently higher age-equivalent scores on the expressive vocabulary than the 2 other groups, which did not differ from each other.

Table 2.

Results of post-hoc LSD tests by group for RV AE and EV AE

Comparison of the gap between mean scores of receptive and expressive vocabulary among the groups

Relative receptive and expressive vocabulary abilities for the three groups (HF ASD, LF ASD, and ID) were compared using mean age-equivalent REVT scores as can be seen in Table 2. Figure 1 shows that the score of age-equivalent expressive vocabulary (EV AE) is higher than the score of age-equivalent receptive vocabulary (RV AE) for the HF ASD and LF ASD groups. However, the score of EV AE is lower than the score of RV AE for children with ID.

Figure 1.

Receptive and expressive vocabulary trends. HF ASD=high functioning autism spectrum disorder; LF ASD=low functioning autism spectrum disorder; ID=intellectual disability.