How parents introduce new words to young children: The influence of development and developmental disorders

Highlights

• We observed parents introduce novel words for novel objects during interactions.

• Parents readily introduced novel words even when child did not use the words.

• Parents used attention-related strategies before, during, and after offering words.

• Parents altered some strategies as children became more verbal.

• Autism spectrum disorder and Down syndrome affected parents’ strategies.

Abstract

This study documents how parents weave new words into on-going interactions with children who are just beginning to speak. Dyads with typically developing toddlers and with young children with autism spectrum disorder and Down syndrome (n = 56, 23, and 29) were observed using a Communication Play Protocol during which parents could use novel words to refer to novel objects. Parents readily introduced both labels and sound words even when their child did not respond expressively or produce the words. Results highlight both how parents act in ways that may facilitate their child's appreciation of the relation between a new word and its referent and how they subtly adjust their actions to suit their child's level of word learning and specific learning challenges.

Introduction

The broad claim that parents can orchestrate interactions in ways that facilitate children's early word learning is well supported and widely accepted (Adamson et al., 2004, Hoff and Naigles, 2002, Pruden et al., 2006). Of particular note are parents’ actions that might help a child who is just on the brink of rapid vocabulary acquisition solve the fundamental problem of mapping words onto the world (Baldwin, 1995) by carefully placing a new word into the child's on-going stream of attention. For example, a parent might make the link between a novel word and referent more transparent by calling attention to an object (e.g., shaking a bell) and then, once attention is caught, emphatically stating its name (“That's a bell.”) or producing a word for the sound it makes (“Ding-a-ling!”). Or the parent might follow the child's lead and produce labels or sound words for objects only after they have entered the child's field of attention.

There is ample evidence that such strategies can help children learn new words. Most of the evidence comes from experiments that control when and how an experimenter introduces a novel word for an unfamiliar object. These studies verify that introducing a novel word when a child is already attending to an object is more effective than doing so after attempting to redirect a child's attention to the object (Dunham et al., 1993, Tomasello and Farrar, 1986). In addition, systematic observations of parent–child interactions with both typically developing children (TD; Simpi and Huttenlocker, 2007, Smith et al., 1988, Tomasello and Farrar, 1986) and young children with autism spectrum disorder (ASD: McDuffie & Yoder, 2010) support the claim that vocabulary acquisition proceeds more quickly when parents use strategies that take the child's attention into account. Moreover, we know from Clark, 2007, Clark, 2010 detailed analyses of five children's conversations that parents will work very hard to ensure that their child is attending to a referent before they introduce a word and that, when they do so, children as young as two years often repeat the new words.

However, systematic observations of what parents do as they introduce new words to young children are surprisingly limited. This gap leaves unclear whether parents do indeed place new words into the child's stream of attention and how this placement might be influenced by developmental changes in language acquisition strategies and by language learning challenges associated with development disorders. To document what parents do during on-going social interactions, we observed how they spontaneously presented different types of novel words (labels and sounds) to children who were at different points of early word learning (pre- and post- 50 words) and who varied markedly in joint attention capacities and word learning skills (TD toddlers, young children with ASD, and young children with Down syndrome [DS]). Such observations can indicate whether parents attempt to align their supportive actions to suit developmental changes in word learning or atypical patterns of joint attention, and if so, how they adjust their actions.

To observe how parents introduce new words during on-going interactions, we needed to control for the child's prior experience with words and their referents. We did so by integrating the venerable experimental practice of using novel words for novel objects (Berko, 1958; see Ratner & Menn, 2000) into semi-naturalistic observations of parent–child communication. The use of novel words for novel objects has been used to probe children's use of various principles to map words to objects (Moore, Angelopoulos, & Bennett, 1999) and verbs (e.g., Golinkoff, Jacquet, Hirsh-Pasek, & Nandakumar, 1996), and the effects of an unfamiliar adult's actions (e.g., pragmatic cues, Tomasello & Akhtar, 1995; attention management strategies; Dunham et al., 1993; and overhearing, Floor & Akhtar, 2006). But this procedure has rarely been used to reveal how parents present new words (cf., Hani, Gonzalez-Barrero, & Nadig, 2013, who explicitly ask the parent to teach two novel words). In the study reported here, we adapted the procedure by providing novel words for novel objects as one of several suggestions parents might pursue as they played with their child using the Communication Play Protocol (CPP; Adamson et al., 2004), which produces semi-naturalistic playful parent–child interactions during activities which encourage social interacting, requesting, and shared commenting.

Of primary interest was what parents did to modulate their child's attention when they introduced a novel word. Thus each time a parent produced an utterance containing a novel word, we systematically coded what she did before, during, and immediately after this utterance. We expected that most parents would readily follow our suggestion to use the novel words, especially in communicative contexts, such as shared commenting, that afford the introduction of new words. Moreover, as has often been observed in studies of parent–child communication (e.g., Bornstein, Tamis-LeMonda, & Haynes, 1999), we expected that they would be quite consistent over time in how they interacted with their child. Moreover, we expected that parents would often act in ways that provided a communicative frame that scaffolded the child's attention to the word/object mapping. Overall, we expected that parents would often create a frame using strategies that sought to gain the child's attention before they produced the novel word (e.g., shaking the to be named object, pointing to it, saying “look here”), that might draw attention to the novel word during its presentation (e.g., emphasizing the word, showing the object), and that would prompt the child's use of the word after they produced it (e.g., asking what the object is called, eliciting a repetition).

We were also interested in how children's approach to word learning might influence how often and how parents introduce new words. One important variation is developmental. As children begin to amass a vocabulary, they become more rapid and skilled word learners (Hollich, Hirsh-Pasek, & Golinkoff, 2000). Thus we expected that during the beginning period of vocabulary acquisition (<50 words), parents would provide more and possibly different input. More specifically, they might be more likely to repeat words when a child's expressive vocabulary was very small. They might also be more apt to present the novel sound word more often than the novel label when a child has a limited expressive vocabulary and to favor the label after the child's vocabulary has begun to expand. This developmental pattern would be consistent with Werner and Kaplan's (1963, p. 103) intriguing discussion of how early in language development children and parents delight in repeating naturalistic onomatopoetic “nonstandard” words that translate perceived noises into patterns that are “linguistically and morphemically stabilized”. Moreover, they might adjust the communicative frame to suit the child's word learning capacities. More specifically, we hypothesized that parents interacting with a child who was just at the cusp of building an expressive vocabulary would make more attempts to capture attention to the link between the novel word and object than when the child was already demonstrating the capacity to learn new words. Thus they might attempt to gain attention to the novel object before presenting the word, draw attention to the word as they present it, and repeat the word more often. In contrast, we anticipated that once a child was amassing a vocabulary, parents would be more likely to prompt the child to use the word after they introduced it.

In addition to documenting how the child's vocabulary level might influence parents’ strategies for introducing new words, we studied what transpired when children had persistent problems with one or more aspects of the mapping problem. The challenge of optimizing word learning is more complex, and arguably more crucial, when a child has a developmental disorder that compromises joint attention skills or hampers the infusion of symbols into interactions (Diesendruck, 2007). Our three-group design allowed us to examine differences between strategies parents used with typically-developing children and children with developmental disorders, but – perhaps more revealing – between strategies parents used when interacting with children with different developmental disorders.

Children with ASD present an especially complex constellation of early communication problems that contribute to difficulties in early language acquisition. Joint attention deficits that are evident before language emerges and predict language concurrently and longitudinally (Bopp & Mirenda, 2011Charman et al., 2003, Toth et al., 2006) may hamper a child's appreciation of the relation between a partner's speech and shared objects. Limited interest in new objects (Adamson, Deckner, & Bakeman, 2010) may limit opportunities to engage a child long enough to present novel words. Yet, there is now substantial evidence that children with ASD can take advantage of adult scaffolds that support their attention to both words and shared objects. Observational studies of parent–child interactions suggest that language outcome is related to how often a child with ASD sustains periods of symbol-infused joint engagement during which a caregiver supports their shared focus (Adamson, Bakeman, Deckner, & Romski, 2009) and how often parents follow-in on the child's current focus of attention (McDuffie & Yoder, 2010). Moreover, a growing experimental literature indicates that children with ASD can take advantage of adult strategies that help overlay words and objects, for example, when an adult follows the child's focus of attention or provides several repetitions of the new word (Luyster & Lord, 2009; see also Hani et al., 2013, Gliga et al., 2012, Parish-Morris et al., 2007, Walton and Ingersold, 2013).

However, it is not yet clear if parents spontaneously provide such supports to children with ASD when presenting them with new words and if they do so more often or differently than parents who are interacting with children who do not have a comparable joint attention deficit. Previous studies suggest that parents of children with ASD provide social cues – such as looking at the relevant target and pointing while labeling an object – similar to those provided by parents interacting with typically developing children (Hani et al., 2013). But there are intriguing hints as well that parents of children with ASD may make subtle adjustments, such as making objects more perceptually salient as they bid for joint attention (Adamson, McArthur, Markov, Dunbar, & Bakeman, 2001) and continuing to use multiple cues to introduce a label even after a child has a substantial vocabulary, a pattern that is not evident in typically developing samples (Hani et al., 2013). In the current study, we anticipated parents might present fewer words to children with ASD, who tend to show less interest in new objects (Adamson et al., 2010) and more difficulty in sustaining joint engagement (Adamson et al., 2009). But we also expected that they would be more likely than parents in the other two groups to actively seek the child's attention to the novel object before introducing a word and to draw attention to the word as they presented it.

Parents of young children with DS may face a different constellation of challenges when they try to introduce new words. One common observation is that young children with DS seem to be less expressively verbal than one might expect given their relatively intact preverbal joint attention skills and joint engagement experiences (Adamson et al., 2009, Sigman and Ruskin, 1999) and their level of comprehension and nonverbal visual cognition (Chapman, 2003). Several factors might account for their relative reticence include articulation problems that might make spontaneous speech unintelligible (Abbeduto & Murphy, 2004) and problems with auditory short term memory that might make learning new words difficult (see, e.g., Mosse & Jarrold, 2011). Given such problems, we anticipated that the attention modulation strategies used before and during the presentation of novel words by parents of children with DS would not differ from those of parents with TD children but they would be more likely to repeat the words to provide the child with additional experience hearing the word and to compensate for possible memory problems.

Finally, although our primary focus was on the parents, we were also interested in whether the parent's presentation of a word resulted in the child's attending to the word. This is difficult to reliably discern during an interaction since the child might take up a word receptively without producing the word. Thus we focused on indications that the child tried to use the new word during the interaction. We coded responds expressively when the child responded vocally with either the actual word or with a nonverbal vocalization right after the parent used the word. More restrictively, we coded produces word when the child actually said the novel word. Overall, we expected that as children became more advanced word learners, they would respond expressively more often and be more likely to produce the novel word, although we anticipated that many toddlers would not produce many of the novel words during our observations. Moreover, we anticipated that the child's responsiveness might be influenced by child diagnosis, parental strategies, and communicative context. More specifically, we hypothesized that responding expressively would be more likely in the TD and DS groups than in the ASD group, regardless of vocabulary level, and that producing the novel word would be less likely in the DS group than in the ASD or TD group. In terms of parental strategies, we hypothesized that children would be more likely to respond expressively when parents prompted for the novel word. Also, in line with the experimental studies cited above, we hypothesized that gaining a child's attention before presenting the word and presenting the word while the child was attending to the novel object would increase the probability of an expressive response. Finally, we hypothesized that communicative context may influence a child's response to new words such that children would produce new words, or at least respond expressively, more during activities like requesting and shared commenting that focus attention on their referents than during activities like social interacting that focus more on interpersonal dynamics.