Vocabulary Composition in Toddlers with ASD: The Longitudinal Development of a Productive Verb Lexicon

Background: Research suggests that verbs are harder for typically developing children (TD) to learn than nouns, perhaps because their referents are often less perceptually available than words for concrete objects (Gentner, 1982). One theory suggests that the reduced perceptual availability of verbs necessitates increased support from social and grammatical cues, which may put children with less access to social cues at a disadvantage (Maguire et al., 2006). Indeed, experimental research suggests that children with autism spectrum disorders (ASD) have relatively intact noun learning abilities (Swensen et al., 2007); moreover, parental report measures indicate that they use proportionately fewer verbs (Parish-Morris et al., 2009). However, because children with ASD appear to learn verbs via some typical strategies (Naigles et al., 2011), the differences in verb vocabulary may result from children with ASD not using verbs as flexibly and comprehensively as TD children. The current study seeks to distinguish these accounts by examining children’s noun and verb use in naturalistic conversations.  

Objectives: Plot a longitudinal trajectory of growth in 2 lexical categories (nouns and verbs) over two years, investigating whether the lexicons of children with ASD have fewer verbs than TD children, and whether this discrepancy increases over time.

Methods: Seventeen toddlers with ASD (mean age=32.86 months) were matched to 18 TD toddlers (mean age=20.60 months) on Expressive and Receptive Language abilities at Visit 1 of 6. The language produced by parents and children during six 30-minute unstructured home-based play sessions (4 months apart) was recorded and transcribed in CHAT format. 

Results: Preliminary analyses revealed that both groups had proportionately fewer verbs than nouns in their early productive lexicons (Bornstein et al., 2004). The proportion of verbs and nouns used by children at each visit was calculated by dividing the number of category tokens by the total number of words spoken. Independent-samples t-tests conducted at each time point revealed that although the TD group used a significantly smaller proportion of verbs than the ASD group at Visit 1, t(32)=-2.10, p=.04, rapid growth in the TD group reversed this pattern at Visits 3, 5, and 6, when they used a significantly larger proportion of verbs than the ASD group, (ps<.05 at Visits 3 and 6, p=.06 at Visit 5). In contrast, TD children used a larger proportion of nouns in their conversation at Visits 2 and 3 (ps<.05) than did children in the ASD group.  Paired-samples t-tests conducted within-groups revealed that the proportion of nouns in the vocabularies of children with ASD did not change between Visits, whereas the proportion of nouns used by the TD group decreased from Visit 2 to Visit 3 (p=.009), from Visit 3 to Visit 4 (p=.06), and from Visit 5 to Visit 6 (p=.03). 

Conclusions: This study demonstrates that toddlers with ASD produce proportionately fewer verb tokens than TD control children at multiple time points, during naturalistic interactions. Future analyses will use growth curves to characterize developmental trends in the vocabulary of each group, and explore how the verb usage of children with ASD differs from TD.