Third person singular -s in typical development and specific language impairment: Input and neighbourhood density

Abstract

The purpose of this study was to examine factors promoting the use of third person singular -s by 23 children with specific language impairment (SLI) and 21 children with typical development (TD). Relative proportions of third person singular -s forms in the input (input proportion) were calculated for 25 verbs based on data from an American English corpus of child-directed speech. Neighbourhood density values were also collected for these verbs. With previously collected probes of third person singular -s use for each of these verbs, we found with logistic regression that input proportion was positively associated with the likelihood of third person singular -s use for both groups. For neighborhood density, we found that children with SLI were more likely to inflect sparse verbs than dense verbs; density was not significantly related to inflection use for TD children. We argue that as a result of their verbs’ poorly encoded phonological representations, children with SLI were less able to inflect dense verbs than sparse verbs. We recommend that clinicians be aware of the effects of input proportion and neighbourhood density to ensure that assessments are representative and that treatment success is optimal.

MOSAIC+: a cross-linguistic model of verb-marking in typically developing children and children with Developmental Language Disorder

“…Our results show that the addition of a defaulting mechanism allows MOSAIC to simulate the very high rate of OI errors in early child English without affecting the model's previously good fit to the data on Dutch, German, and Spanish. These findings are consistent with the idea that at least some apparent OI errors in English reflect a process of defaulting to the most frequent form of the verb (Räsänen et al, 2014;Kueser et al, 2018), and suggest that the very high rate of OI errors in English reflects the fact that, in English, but not in the other languages, defaulting tends to result in the same kind of errors as the learning of infinitives directly from the input.…”

“…However, the pattern of verb-marking deficit in DLD varies across languages. Thus, as we have already seen, English-speaking children with DLD tend to produce OI/bare stem errors at higher rates than language-matched controls, even at high MLUs (Kueser et al, 2018;Rice, Wexler, & Cleave, 1995;Rice, Wexler, & Hershberger, 1998). However, this effect appears to be specific to English.…”

“…This effect is difficult to explain under the assumption that all English-speaking children's uses of zero-marked forms in 3sg contexts are OI errors, and suggests that at least some of English-speaking children's verbmarking errors reflect a process of defaulting to the bare stem. Interestingly, this result has since been replicated by Kueser, Leonard, and Deevy (2018) in a group of English-speaking children with DLD (with an average MLU of 4.08) and a group of MLU-matched controls. In the Kueser et al study, although the children with DLD made significantly more errors than the TD children, both groups showed a significant input effect of the kind reported by Räsänen et al (2014).…”

Simulating the Acquisition of Verb Inflection in Typically Developing Children and Children With Developmental Language Disorder in English and Spanish

“…Evidence for defaulting effects in English DLD is provided by Kueser, Leonard, and Deevy (2017), who replicate the findings of Räsänen et al, but find that children with DLD produce more bare forms than typically developing children, thus raising the possibility that defaulting effects are stronger in children with DLD. Bedore and Leonard (2001) find that Spanish children with DLD do produce more errors than age‐matched (but not MLU‐matched) controls.…”

Phonological Treatment Options for Children with Expressive Language Impairment

“…34 Children with DLD are also affected by the same phonological variable but show the opposite pattern. Specifically, children with DLD produce the third person singular morpheme more accurately in verbs that are more phonologically distinct from other words in the language, both on elicited probes 35 and in the context of treatment. 36 Children with DLD are also more accurate in their production of regular past tense when the verbs to be inflected are composed of common sound sequences.…”