photo blocks

Who did we study and why?

Late language emergence (LLE) is a very common developmental disorder among toddlers (Rescorla, 1989). In brief, LLE is defined as a delay in language learning that is not caused by an underlying cognitive or motor issue. Common symptoms of LLE include small expressive vocabularies and utterances with less grammatical complexity (Rescorla, 1989; Thal et al., 2013).

Previous studies have supported the importance of parent input in child language development. Both the quality and the quantity of the language input can affect how the child develops language (Bernstein Ratner, 2013). Since parent input is so important, I was interested in learning about how parents speak to their children with typical development (TD) and LLE. Do parents speak differently to their children based on whether they have a child with TD or LLE? If so, how?

What aspects of language did we study?

I decided to specifically look at how parent language differed between groups based on recasting and grammatical complexity. Recasting is when a parent repeats their child’s utterance in a way that adds information to the original utterance without changing the intended meaning (Cleave et al., 2015). For example, if a child says “I no see the gooses,” their parent might respond by saying, “Oh no, you didn’t see the geese?” In this example, the parent added the auxiliary verb “do” and changed the plural of the noun “goose.” Overall, recasting has been shown to be beneficial in language intervention and is important for parent use in conversations with their children (Cleave et al., 2015; Che et al., 2018).

As for grammatical complexity, a common finding in the literature is that more complex parent language correlates with more complex child language (Huttenlocher et al., 2002). Overly simplified language takes away aspects of morphology and syntax that are crucial for children to recognize (van Kleeck et al., 2010). Thus, it is important for parents to use rich grammatical utterances with their children.

What does the literature say?

According to the “fine-tuning” hypothesis, it is possible that parents change the way they speak to their child based on the child’s language abilities (Snow & Ferguson, 1977). For example, if a child uses language that is less grammatically complex, the parent will adjust their language to be less complex as well. Likewise, if a child uses very complex language, the parent will accommodate their child by using more complex language. By definition, children with LLE tend to have lower levels of complexity in their language output. This means that the “fine-tuning” hypothesis would predict parents of children with LLE to use less complex language with their children than parents of children with TD.

In terms of grammatical complexity and recasting, the “fine-tuning” hypothesis is somewhat supported by the literature. One study showed that parents of children with TD are more likely to respond to their child than parents of children with LLE (Vigil et al., 2005). More responses indicate a greater chance for more complex interactions. A different study showed that parents of children with TD are more likely to use complex recasts than parents of children with LLE, although the frequency of recasting was not shown to differ between groups (Conti-Ramsden, 1990; Paul & Elwood, 1991). Finally, one study did not find any syntactic differences between
parents of children with TD and parents of children with LLE (Paul & Elwood, 1991). The divided findings and lack of longitudinal research on this subject indicate the need for further research. The work we did aimed to add longitudinal evidence to the current literature in order to better understand the interaction between parent and child language.

The current study

We used CHILDES, a publicly accessible database of child language transcripts, in order to complete this study (see Ellis Weismer et al., 2013). The language transcripts used in the study included only parent-child play sessions.

36 children with LLE and 23 children with TD were analyzed at the ages of 30 and 42 months. Children were defined as having LLE if they were below the 10th percentile on the MacArthur CDI: Words and Sentences at 24 months. Furthermore, participants were matched based on nonverbal cognition and maternal education.

The Computerized Language ANalysis Program (CLAN; MacWhinney, 2000) was used to analyze the language transcripts. In total, we analyzed 12,395 child utterances and 23,054 mother utterances. Percent overlapping utterances (POU; the percent of parental utterances that included words from their children’s preceding utterance) was used as our operationalized definition of recasting and was computed using CHIP (Sokolov & MacWhinney, 1990). MLU in morphemes was used as our operationalized definition of grammatical complexity.

What did we find?

At 30 months, we found that parents of children with LLE used less grammatically complex language when playing with their children than parents of children with TD. However, this difference no longer existed by the time the children were 42-months old. This suggests that parents of children with LLE increase their use of grammatically complex language as their children age, whereas parents of children with TD consistently use grammatically complex language throughout their child’s development.

As for recasting, we found that at both time points, parents of children with TD recasted more than parents of children with LLE. This finding was rather disappointing, as children with LLE could surely benefit from being exposed to more recasts. It is unclear why parents of children with LLE recast less, however, one explanation is that parents of children with LLE have less of an opportunity to recast because their children speak less than the children with TD. However, after a more in-depth statistical analysis of our results, it seems that parents of children with LLE might recast less despite the quantity of child utterances. Future research is necessary to
confirm this speculation.

These results support the literature claiming that there are differences in parent language between groups of children with TD and LLE. This work adds to our current understanding of parent-child interactions by providing information from longitudinal data. Clinically, speech-language pathologists can use this information to encourage families affected by LLE to consider recasting more frequently with their children.


Bernstein Ratner, N. (2013). Why talk with children matters: Clinical implications of infant- and child- directed speech. Seminars in Speech and Language, 34 (4), 203-214. doi: .

Che, E, S., Brooks, P, J., Alarcon, M, F., Yannaco, F, D. (2018). Assessing the impact of conversational overlap in content on child language growth. Journal of Child Language, 45 (1), 72-96. doi:

Cleave, P, L., Becker, S, D., Curran, M, K., Owen Van Horne, A, J., & Fey, M, E. (2015). The efficacy of recasts in language intervention: A systematic review and meta-analysis. American Journal of Speech Language Pathology, 24 (2), 237-255. doi: 10.1044/2015_AJSLP-14-0105

Conti-Ramsden, G. (1990). Maternal recasts and other contingent replies to language-impaired children. Journal of Speech and Hearing Disorders, 55, 262-274. doi: 10.1044/jshd.5502.262

Ellis Weismer, S., Venker, C., Evans, J. L., & Moyle, M. (2013). Fast mapping in late-talking toddlers. Applied Psycholinguistics, 34, 69-89. doi: 10.1017/S0142716411000610

Huttenlocher, J., Vasilyeva, M., Cymerman, E., Levine, S. (2002). Language input and child syntax. Cognitive Psychology, 45 (3), 337–374.

MacWhinney, B. (2000). The CHILDES project: Tools for analyzing talk (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates.

Paul, R., & Elwood, T, J. (1991). Maternal linguistic input to toddlers with slow expressive language development. Journal of Speech and Hearing Research, 34(5), 982-988. Rescorla, L. (1989). The language development survey: A screening tool for delayed language in toddlers. Journal of Speech and Hearing Disorders, 54, 587-599. doi: 10.1044/jshd.5404.587

Rescorla, L. (1989). The language development survey. Journal of Speech and Hearing Disorders, 54 (4), 587-589. doi:

Snow, C, E., & Ferguson, C, A. (1977). Talking to children: Language input & acquisition. Cambridge: Cambridge University Press.

Sokolov, J. L., & MacWhinney, B. (1990). The CHIP framework: Automatic coding and analysis of parent-child conversational interaction. Behavior Research Methods, Instruments, & Computers, 22 (2), 151-161. doi: 10.3758/BF03203138

Thal, D. J., Marchman, V. A., & Tomblin, J. B. (2013). Late-talking toddlers: Characterization and prediction of continued delay. In L. A. Rescorla & P. S. Dale (Eds.), Late talkers: Language development, interventions, and outcomes (pp. 169-201). Baltimore, MD: Brookes.

van Kleeck, A., Schwarz, A.L., Fey, M., Kaiser, A., Miller, J., Weitzman, E. (2010). Should we use telegraphic or grammatical input in the early stages of language development with children who have language impairments? A meta-analysis of the research and expert opinion. American Journal of Speech-Language Pathology, 19 (1), 3–21.

Vigil, D, C., Hodges, J., & Klee, T. (2005). Quantity and quality of parental language input to late-talking toddlers during play. Child Language Teaching and Therapy, 21 (2), 107-122. doi: 10.1 191/0265659005ct284oa

Sarah Elazar completed this work under the direction and guidance of Dr. Nan Bernstein Ratner and Julianne Garbarino. This research was presented at the 2019 American Speech Language and Hearing Association Convention in Orlando, Florida. Sarah graduated with her BA in May of 2020.

Expenses related to this research were funded by the MCM Fund for Student Research Excellence. This award is designed to support independent student research projects, and is made possible by an anonymous donation to the Department and by other donations by faculty, alumni, and friends. Please consider donating to the fund through the following link: