Within the wide range of techniques we speech and language pathologists (SLPs) carry in our speech-language toolboxes, the use of recasts is one we pull out frequently. In fact, it’s a tool that many SLPs, parents, and teachers often use intuitively and with ease when interacting with young children. But how valuable, exactly, is the use of recasting? What research guides the use of this technique in practice? And when do we need to use something more than recasts to help language grow? Recent research provides some clues to help us to answer these questions.
Before taking a look at that research, though, we first need to define what a recast is. A recast occurs when an adult responds to something a child says by repeating the child’s words while correcting the child’s errors. Here’s a quick example of a recast in use:
Child: “Daddy home”
Adult: “Yes! Daddy’s home!”
There are a couple of things that are really important to the definition of a recast. First, notice how, in the example above, the child started the interaction and the adult responded to him. This is an essential element of a recast. Note, too, how the child isn’t required to or expected to imitate the adult sentence. Recasts are used conversationally, as a way of providing a child with an indirect model, or a casual example, of the language that he should have used in his sentence. We believe that this helps a child to compare what he said to what the adult said - not consciously, per se, but as a way of figuring out how the language system works. While all this is happening, though, it should simply feel like an interactive conversation to the child. Another thing to note is that the adult in our above example kept the child’s words and the meaning of the child’s sentence the same. This makes recasts (also sometimes called expansions), different from extensions. Extensions are used to add onto the topic of conversation, rather than add in missing or incorrect information. If the adult in the example above had said “Yes, he’s coming inside now,” it would have been an extension, rather than a recast or expansion.
There’s one more important detail about recasts to discuss before we look at their effectiveness. Recasts can be either broad or focused. Recasts are considered to be broad when an adult responds to anything a child says with a recast that corrects errors or adds in missing information, across a broad range of different speech and language areas. The adult might add new vocabulary, might model better speech sounds, or might correct the grammar of a child’s sentence, depending on what the child said. If a child says “Want it,” an adult might say “You want juice?”; if a child says “I want a poon,” an adult might say “You want the ssspoon?”; if the child says “Daddy eating,” the adult could say “Daddy is eating!” With focused recasts, though, the adult focuses on providing examples of a very specific part of language. Say, for example, the child needed to work on using the articles “a” and “the” in his language. An adult would focus on making sure recasts contained “a” and “the” repeatedly and frequently. If the child said, “I see car,” the adult would say, “I see the car too!”; if the child said “I want spoon,” the adult would say, “You want a spoon?” And so on. The difference between broad and focused recasts is important, because they appear to have a different impact on language learning for children with language delays and/or impairments.
Research hasn’t yet given us all the answers about how and when to use recasts. However, there do seem to be some guiding principles that have gained consensus. Recasts seem to be especially effective for young children who need work on the syntactic, or grammatical, aspects of their language. It also appears that it is important to use focused, rather than broad recasts (Cleave et al, 2015). This means that recasts may work best when a specific grammatical target has been chosen for the child and everyone in that child’s life is focused on using recasts that include that element of grammar. For toddlers and preschoolers, this might involve grammatical morphemes, or word parts, such as present progressive –ing (swimming), plural –s (cats), possessive –s (mom’s shoe), past tense –ed (jumped), the words “is” and “are” used in sentences (“She’s a teacher; I’m eating an apple; “Is dad home?”), or pronouns (He is painting). The adults in the child’s life might pick one or two targets such as these and work to make sure that their recasts contain those targets.
The rate at which we use recasts also seems to be pretty important. Even though many people tend to use recasts intuitively, we might not be using this technique enough to really make a difference. Research indicates that, to lead to changes in a child’s language, recasts need to be used at a rate of 1-2 times per minute. This is 2-3 times the rate at which parents use recasts naturally (Cleave, et al., 2015; Eisenburg, 2014, Fey, Long & Finestack, 2003), which means that we might really need to step up our recasting game if we are really going to effect change in a child’s language system. The good news is that recasts don’t need to be provided just by SLPs. Research indicates that parents and teachers can be effective in providing the right kind and right number of recasts as well, especially when provided with training (Cleave et al., 2015; Fey, Long & Finestack, 2003). This is important, because it’s best for children to hear examples of language targets across people, settings, and time.
The population with which recasts are used may also be an important variable. The evidence behind the use of recasts is the most strong for children who have specific language impairments; in other words, those children who have difficulty with language, but who don’t have another diagnosis such as Down Syndrome, Hearing Loss, or Autism Spectrum Disorder. The jury is still out on the effectiveness of only using recasts (or other input-based approaches) with kiddos who have a more substantial developmental delay. This isn’t to say that recasts are never effective with these children. However, for children with diagnoses other than a specific language impairment, we may need to use strategies that are focused on production as well as input (Cleave, et al., 2015).
We don’t have all the answers about when, how, and with whom to use recasts. But, we do have a beginning roadmap to help us grow the language of toddlers and preschoolers who struggle with the grammatical aspects of language development. Armed with this knowledge, we can make an important difference in the language systems of these children.
Becca Jarzynski, M.S., CCC-SLP is a pediatric speech-language pathologist and a Clinical Instructor at UW- Eau Claire. You can find more of her blog posts at www.talkingkids.org.
Cleave, P., Becker, S., Curran, M., Owen Van Home, A. & Fey, M. (2015). The Efficacy of Recasts in Language Intervention: A Systematic Review and Meta-Analysis. American Journal of Speech-Language Pathology, 24, 237-255.
Eisenburg, S. (2014). What Works in Therapy: Further Thoughts on Improving Clinical Practice for Children with Language Disorders. Language, Speech, & Hearing Services in Schools, 45, 117-126.
Fey, M., Long, S., & Finestack, L. Ten Principles of Grammar Facilitation for Children with Specific Language Impairments. (2003). American Journal of Speech Language Pathology, 12, 3-15.
Yoder, P., Woynaroski, T., Fey, M., E., & Warren, S.F. (2014). Effects of dose frequency of early communication intervention in young children with and without Down Syndrome. American Journal on Intellectual and Developmental Disabilities, 119, 17-32.