Bidirectional language influence in mother–child interaction and its effects on the communicative development of children with cochlear implants: A longitudinal study

Abstract

The present study aimed to investigate the communicative characteristics of children with cochlear implants (CIs) and their mothers in interaction, whether and how they differ from those of mother–child dyads with normal hearing, and whether mother and child influence each other over the first year after implantation. Eighteen Italian-speaking children with CIs were assessed longitudinally across four time points, from before implantation (ages: 10–33 months) to 1 year after implantation. Two groups of normally-hearing children matched for age or lexical production level were used as comparison groups. The language used by mothers in semi-structured play sessions with their children was analyzed for use of specific techniques aimed at eliciting language in children with CIs (or Facilitating Language Techniques [FLTs]: questions, labeling, comments, parallel talk, reformulations, and directives); children’s language was analyzed for communicative functions (initiatives and adequate responses). Results showed delayed communicative development for children with CIs with respect to both control groups and different use of FLTs such as directives between mothers of children with or without normal hearing. Findings also highlighted an effect of the interlocutor’s communicative behavior for both children and mothers: In the group of children with CIs, children’s initiatives were hindered by a higher use of questions; mothers’ labeling was negatively affected by children’s responsivity. Findings may have important implications for language development support programs, families, and professionals.

Keywords

Cochlear implants children maternal input communicative development facilitating language techniques

Introduction

After cochlear implantation, the speech perception and language skills of children who are deaf and hard of hearing (DHH) undergo substantial improvements (Quittner et al., 2004; Spencer & Marschark, 2010; Uhler et al., 2011). Once children with cochlear implants (CIs) gain access to sound, they acquire words following a trajectory that appears to resemble that shown by children with normal hearing (henceforth, NH) in the first year of life (Li et al., 2020). However, children with CIs have the opportunity to learn spoken language only when they are at least 1 year older than their age-matched peers with NH; as a result, the expressive vocabulary skills (Lund, 2016; Nittrouer et al., 2018; but see also Wie et al., 2020) and the conversational abilities (Szarkowski et al., 2020) that they develop are comparatively reduced. One year after implantation, children with CIs still produce fewer words and fewer different words than their age-matched peers with NH (Kondaurova et al., 2020; Majorano et al., 2020). The developmental trajectory of communicative functions is less clear: In the Italian context, Rinaldi et al. (2013) report that children with CIs tested between 7 and 25 months post-implantation are less able to initiate topics and make requests than age-matched peers with NH; whereas Lavelli et al. (2018) report comparable proportions of initiatives and adequate responses in children with NH and in children with CIs tested between 12 and 38 months post-implantation.

Besides being related to individual factors (Caselli et al., 2012; Yoshinaga-Itano et al., 2017), outcome variability in children with NH and especially in children with CIs appears to be related to environmental factors such as parental input quantity and quality (Holzinger et al., 2020; Lavelli et al., 2018; Majorano et al., 2018; Rowe et al., 2017; Weizman & Snow, 2001). In the literature, parental input has been investigated by looking at a wide range of parent behaviors, including how they gain and maintain their child’s attention, how sensitive they are in terms of emotional warmth, availability, responsivity, and consistency, how they increase their child’s access to language (e.g. by being face to face and closer to the amplification device and by exaggerating visual and vocal cues), and by looking at the quantity and quality of the sign or spoken input they provide (see Curtin et al., 2021, for a systematic review of studies assessing parent-child interaction with DHH children). The present work focuses on spoken language input quality, which has most often been studied in previous research by looking at the techniques used to elicit language from children (DesJardin et al., 2007, 2009, 2014; Dirks et al., 2020).

According to DesJardin and colleagues, the techniques that ‘facilitate language’ in DHH children (or ‘Facilitating Language Techniques’; FLTs) include questions, labeling (i.e. naming surrounding objects), comments, parallel talk (i.e. describing the child’s ongoing actions), reformulations (the repetition of a child’s utterance or the expansion of that utterance to contain new syntactic and/or semantic material), and directives. Most of these techniques have been found to be associated with CI users’ verbal comprehension and expressive language skills (Cruz et al., 2013; DesJardin & Eisenberg, 2007), in some cases independently of age at implantation (Szagun & Stumper, 2012), a well-known factor affecting language acquisition in these children (Geers et al., 2009). However, some techniques seem to elicit language less efficiently than others (e.g. comments as compared with questions; Justice et al., 2002; Tompkins et al., 2013) or only when children are still young and linguistically inexperienced (e.g. labeling; Yoder et al., 2001). Others even seem to have hindering effects: Although Desjardin and colleagues categorize them as FLT, directives are negatively associated with children’s language outcomes (DesJardin & Eisenberg, 2007; Nelson, 1973), especially when parents redirect the attention of the child toward objects or activities that are not the center of his focus (Akhtar et al., 1991).

There are mixed reports on whether mothers of children with CIs use more or fewer of these techniques. Because parents modulate their input quantity and quality in relation to the children’s actual or perceived language abilities (Farran et al., 2009; Vanormelingen et al., 2015) and because children with CIs lag behind their peers with NH in a number of areas, including vocabulary (Luckner & Cooke, 2010), phonology (Flipsen & Parker, 2008), and morphosyntax, when they speak a language that is rich in its morphosyntactic repertoire such as Italian (Caselli et al., 2012; Guasti et al., 2014), differences between groups should be expected. This should be especially valid in the early years of the child’s life, when differences in lexical knowledge between children with CIs or NH are largest (Fulcher et al., 2012). Yet, findings are varied. Ambrose et al. (2015) showed that 3-year-old children with mild to severe deafness wearing hearing aids (HAs) are exposed to a lower proportion of conversational-eliciting utterances than children with NH, arguably because their mothers are aware of their limited language skills. DesJardin et al. (2014) showed that parents of children with mild to severe deafness wearing HAs produce more labeling and comments than parents of children with NH and comparable proportions of higher level FLTs such as open questions. In a recent study on children with moderate deafness wearing HAs or with NH and their parents, Dirks et al. (2020) showed comparable proportions of directives between groups but a lower production of higher level FLTs (conversational-eliciting utterances and genuine requests for information, in their study) for the former group. Finally, Wang et al. (2020) found comparable proportions of maternal repetitions (included here under the umbrella term reformulations) in mothers of children with profound deafness wearing CIs and of hearing or chronological age-matched children with NH, which suggests that proportions are not affected by child hearing status.

In summary, research studies on communicative functions in DHH children and on the use of FLTs by their mothers report contrasting results. It is possible that discrepancies between findings depend on the other speaker’s communicative behavior being overlooked. As argued by Lavelli et al. (2018), mothers’ communicative dynamics are always influenced by the child’s communicative behavior (and vice versa): Mothers may talk more or use more reformulations in an attempt to fill the silence created by the absence of a response from the child, and these techniques may, in turn, foster or hinder language development. However, to our knowledge, no study has investigated both the effect of maternal input on the language of children with CIs over the first year after implantation and that of child language on maternal input in interaction. Discrepancies may also depend on the degree of children’s deafness (going from mild to profound in the above-mentioned studies; for its effect on language outcomes, see e.g. Cupples et al., 2018), on the compared NH group, and/or the specific time point observed: Very few studies are longitudinal, focus on children with severe to profound deafness wearing CIs, and include both age-matched and language-matched children with NH as comparison groups.

The present study

This study investigated the spontaneous language production of children with CIs and of their mothers in interaction across four time points, the day before surgery (T1) and 3 (T2), 6 (T3), and 12 (T4) months after implantation. Two control groups of children with NH, one age-matched, the other language-matched, were involved. All children were in the process of acquiring Italian, a language that has received comparatively less attention than English (Guasti et al., 2014; Majorano et al., 2018), but that could lead to wider between-group lexical differences in children due to its richer morphosyntactic system.

The first aim of this study was to investigate whether children with CIs acquiring Italian show delayed communicative development over the first year after implantation (Aim 1). We expected children with CIs to produce fewer initiatives and adequate responses than their age-matched NH peers and to have a performance mostly in line with that of younger, language-matched NH peers. In line with their language quantity and quality developmental trajectories (reported in Majorano et al., 2020, 2021), we further expected the children with CIs tested to significantly improve their communicative behavior over time, but to still lag behind their age-matched NH peers 12 months after implantation.

The second aim of the study was to investigate maternal input in interaction with children with CIs over the first 12 months after implantation and to compare it with that received by children with NH (Aim 2). In the present study, we were especially interested in mothers’ use of FLTs (following the definition of Desjardin and Eisenberg, 2007) and, in particular, in their use of questions, labeling, comments, parallel talk, reformulations, and directives. While we had no predictions regarding mean proportions of use across groups, we expected differences in maternal use of FLTs for the group of children with CIs between pre- and post-implantation, in line with previous studies showing how mothers of children with CIs adapt their speech to the child’s hearing – rather than chronological – age (Bergeson et al., 2006; Kondaurova & Bergeson, 2011).

The third aim of the study was to examine whether mothers’ use of FLTs affected two functions of children’s verbal interactive behaviors, that is, initiatives and adequate responses, and whether these effects changed over the first 12 months after implantation (Aim 3). In line with previous studies on children with CIs (e.g. Lavelli et al., 2018), we expected maternal communicative behavior to influence and predict child communicative dynamics.

The fourth aim was to assess whether children’s verbal communicative behavior influenced their mothers’ production of FLTs (Aim 4). Based on the studies on coordination dynamics in interaction (e.g. Smith & McMurray, 2018), and complementarily to Aim 3, we expected maternal behavior to be influenced by child behavior.

The present study extends previous work with Italian children (Lavelli et al., 2018; Majorano et al., 2018, 2020) (1) to a younger group of children, (2) to include the comparison of these characteristics pre- and post-implantation, and (3) to more thoroughly investigate bidirectional language influences. With respect to Majorano et al. (2020), which showed delayed language development for children with CIs (in line with previous studies, for example, Fagan et al., 2014; Kondaurova et al., 2020), the present work also includes an analysis of the communicative functions used by the same children and of their effect on maternal communicative behavior (and vice versa).

Method

Participants

Eighteen children with CIs (G-CI group; nine males) were recruited from the ‘Guglielmo da Saliceto’ hospital and matched with 40 children with NH, 20 of the same chronological age (G-NHA group; eight males; age range at T1 = 10–32 months), t(17) = 0.325, p = .749, and 20 with similar, t(17) = 0.772, p = .451, language production abilities at T2 (G-NHL group; 9 males; age range at T1 = 11–12 months), as measured by standardized tests (the Italian version of the MacArthur-Bates Communicative Development Inventory: MB-CDI, Caselli et al., 2015). All the children with NH were recruited from early education centers in northern Italy. See Table 1 for more information on the participants.

Table 1.

Participant characteristics.

	Children with CIs (n = 18)		Age-matched NH group (n = 18)		Language-matched NH group (n = 18)
	M (SD)		M (SD)		M (SD)
Age at diagnosis (months)	6.94	(8.43)	/	/	/	/
Age at implantation (months)	16.83	(7.53)	/	/	/	/
PTA (dB/HL)	100.56	(5.39)	/	/	/	/
Age (months)
T1	16.79	(7.56)	16.73	(7.23)	11.29	(0.41)
T2	21.70	(7.54)	20.65	(7.49)	15.26	(0.29)
T3	24.72	(7.63)	24.36	(7.99)	19.56	(2.16)
T4	30.26	(7.39)	30.70	(8.58)	25.23	(2.68)
Vocabulary production scores
T1	0.56	(1.38)	21	(34.26)	0.94	(2.21)
T2	5.94	(8.95)	33.11	(35.84)	5.44	(6.83)
T3	12.11	(22.31)	38.78	(34.93)	19.50	(20.35)
T4	28	(26.17)	76.61	(28.01)	56.33	(29.72)
Mothers’ education (years of study)	14.00	(5.10)	15.94	(2.80)	17.69	(3.03)

Note. The vocabulary production scores at T2 used to match participants were measured using the MB-CDI test. PTA: Pure Tone Average test performance (250–500–1000–2000–4000 Hz); T1: first session (for the G-CI group, before implantation); T2: second session (for the G-CI group, 3 months after implantation); T3: third session (for the G-CI group, 6 months after implantation); T4: fourth session (for the G-CI group, 12 months after implantation); NH: normal hearing; SD: standard deviation; G-CI: children with CIs; MB-CDI: MacArthur-Bates Communicative Development Inventory; CIs: cochlear implants.

To be enrolled in the study, each child had to be free of diagnosed sensorimotor or developmental disorders and cognitive disabilities. Each deaf child had to have received a diagnosis of severe to profound bilateral sensorineural deafness (> 71 dB, Clark, 1981) and CI surgery by 36 months of age, have hearing parents, participate in the same auditory-verbal therapy rehabilitation program pre- and post-implantation once a week, and have received no exposure to any other language, including sign language. Note that attendance to the auditory-verbal therapy sessions was the same for all DHH children (100%). Children in the control groups had to have reported normal bilateral hearing.

Demographic information on the children and families was collected using the background information form included in the MB-CDI test. All participating children and mothers were Italian monolingual speakers and were of Italian nationality. Mothers’ years of education differed between groups, F(2, 50) = 8.111, p = .006, and were therefore accounted for in the analyses. See Table 1 for more information.

Each family signed an informed consent form; the data were treated in accordance with the privacy regulations (Legislative Decree 196/2003). The protocol for the study was approved by the institutional local ethical committee (protocol number 2017/46268).

Procedure

All dyads were recruited through letters sent to the children’s families in which mothers were informed about the study and asked to participate together with their 10- to 36-month-old children. All children were recruited around the same time; the children with NH were individually matched with the children with CIs at the end of the study based on their chronological age at T1 (G-NHA group) or on their lexical level at T2 (G-NHL group).

All participants were observed four times over a period of a year. At each time point (or session), mother–child dyads were video-recorded during semi-structured play in the laboratory, following the Assessing Linguistic Behavior protocol (Olswang et al., 1987). All sessions were conducted by the same experimenter, that is, the second author (a doctoral student in Developmental Psychology at the time of the data collection) and carried out with the use of the same materials. At each session, four sets of different toys were presented, 1 every 5 minutes, to elicit interactions and spontaneous communication between mother and child (see Majorano et al., 2020, for more details). Mothers were instructed to play with their children as they usually did at home. All interactions were preceded by a few activities for familiarization purposes and were recorded for approximately 20 minutes using the same video camera with a built-in 5.1-channel surround microphone. This time range is in line with that of previous studies on mother–child interaction with children with CIs/NH (Kondaurova et al., 2020; Lavelli et al., 2018; Majorano et al., 2018). The quality of the recordings was verified by the experimenter after each session. Each mother completed the MB-CDI test at each session.

Coding

All video observations were transcribed in their entirety following the Codes for the Human Analysis of Transcripts (CHAT) conventions within the Child Language Data Exchange System (CHILDES; MacWhinney, 2000). The words (content and function words and interjections) produced by mothers and children were identified following the recommendations of Vihman and McCune (1994) and by looking at criteria based on context, vocalization shape, and relation to other vocalizations. Non-speech sounds (e.g. laughter and sneezes) were transcribed but not coded. Utterances were identified following Moneglia and Cresti’s (1997) criteria as units representing a comprehensive intent as indicated by intonation and/or a pause.

To analyze FLTs, all maternal utterances were coded into seven mutually exclusive categories of communicative functions adapted from DesJardin and Eisenberg (2007) and from Longobardi (1992). The categories considered for analysis were questions, labeling, comments, parallel talk, reformulations, and directives (see Table 5 in the Appendix 1 for more details). The utterances that did not belong to any of these (n = 2,345; 3.4% of the total) were coded as ‘other’ and not further analyzed. The proportion of each FLT was calculated by counting the number of occurrences of each on the total number of utterances (FLTs and ‘other’) produced per minute at each session.

Children’s communicative development was analyzed adapting the categories of communicative functions used in Lavelli et al. (2018). For the present study, we only selected and coded the utterances that could be identified as initiatives or adequate responses. Responses were considered ‘adequate’ if they were semantically and/or linguistically related to the mothers’ question (see Table 5 for an example sentence). In this case, vocalizations were also considered as long as they were produced in the context of joint attention and recognized by the mother as clear initiatives or adequate responses.

Interrater reliability was calculated between two independent coders (the second author, who had extensive experience in transcription and coding methods, and an undergraduate student in Educational Science trained by the second author) on 20% of the participants. Interrater agreement on the transcriptions was 92% for mothers and 90% for children. Coding reliability was calculated as Cohen’s kappa. The average kappa was 0.90 for mothers and 0.88 for children.

Data analysis

Children’s and mothers’ communicative characteristics were investigated using Linear Mixed-Effects Models (LMMs) in R (R Development Core Team, 2016; ‘lmer’ function in the ‘lme4’ package, Bates et al., 2015). Each base model included Group and Session as fixed effects, Maternal education (years of study) as covariate, and Subject as random effect. Post hoc tests were run, after main effects, with the Tukey correction (‘glht’ function in package ‘multcomp’; Hothorn et al., 2008), and, after significant interactions, with the Holm correction (‘testInteractions’ function in package ‘phia’; De Rosario-Martinez, 2015). These tests addressed the question of whether children’s (Aim 1) and mothers’ (Aim 2) communicative characteristics differed between groups and sessions.

However, the communicative characteristics of the language produced by children and mothers may not only be influenced by the child’s ability to hear and by age and amount of experience and/or language skills; they may also be influenced by the communicative behavior of their interlocutor. To investigate the hypothesis that mothers’ use of FLTs influenced children’s production of initiatives and adequate responses over and above the effects of group and session (Aim 3), we ran a series of Likelihood Ratio Tests (LRTs). LRTs were run to see whether adding the other speaker’s characteristics (and their interactions with the other predictors) as additional predictors in the base model improved the ability of the model to fit the observed data (or its ‘goodness of fit’). If adding one specific FLT improved the goodness of fit of the model on children’s initiatives or adequate responses, LMMs were updated to include the additional predictor and run again. Because we were especially interested in the G-CI group, post hoc tests with the Holm correction were carried out only in case of significant interactions relative to the additional predictor and Group or both Group and Session. The aim here was to investigate whether the slopes of the relationship between the added predictor and the dependent variable were significantly different between groups. The same approach was used to investigate the hypothesis that children’s communicative behavior influenced their mothers’ use of FLTs over and above the effects of group and session (Aim 4).

Results

Results are reported by aim. For the sake of brevity, in case of multiple significant effects or interactions, we only report the highest level interaction and its relative follow-up tests. In the models including the other speaker’s characteristics as predictors, we only report main effects or interactions with Group.

Aim 1: child communicative development over the first 12 months after implantation

Analysis showed significant interactions between Group and Session for both initiatives and adequate responses (see Table 2). Separate follow-up pairwise comparisons showed that G-CI children produced more initiatives (but not more responses) at T4 than at the previous sessions and fewer responses than both control groups at T4. See Table 2 for more details and Figure 1 to see the children’s trajectories across groups and sessions (adjusted means).

Table 2.

Significant Group × Session interactions and pairwise comparisons relative to the G-CI group.

Subject	Category	F(6, 147)	p	Contrast	Fixed	Value	df	χ²	p
Child	Initiatives	3.936	.001	T1 vs. T4	G-CI	–4.497	1	75.972	<.001
				T2 vs. T4		–4.042	1	61.360	<.001
				T3 vs. T4		–3.411	1	43.708	<.001
	Adequate responses	2.959	.009	G-CI vs. G-NHA	T4	–1.167	1	14.468	.002
	Adequate responses	2.959	.009	G-CI vs. G-NHL	T4	–0.983	1	9.145	.025
Mother	Questions	4.381	<.001	G-CI vs. G-NHL	T3	–2.453	1	12.388	.005
	Questions	4.381	<.001	T1 vs. T4	G-CI	–1.856	1	18.794	<.001
	Labeling	6.210	<.001	G-CI vs. G-NHA	T2	1.115	1	8.424	.030
					T3	1.832	1	22.735	<.001
					T4	2.112	1	30.234	<.001
				G-CI vs. G-NHL	T1	1.094	1	7.259	.049
				G-CI vs. G-NHL	T4	2.194	1	29.182	<.001
				T1 vs. T3	G-CI	–1.161	1	13.139	.004
				T1 vs. T4	G-CI	–1.033	1	10.407	.016
	Comments	12.826	<.001	G-CI vs. G-NHL	T4	4.387	1	40.490	<.001
				T1 vs. T4	G-CI	–5.183	1	115.107	<.001
				T2 vs. T4		–5.158	1	113.999	<.001
				T3 vs. T4		–4.569	1	89.456	<.001
	Parallel talk	8.230	<.001	G-CI vs. G-NHA	T1	–0.390	1	14.848	.001
				G-CI vs. G-NHL	T1	0.470	1	19.528	<.001
				T1 vs. T4	G-CI	0.475	1	26.169	<.001
				T2 vs. T4	G-CI	0.286	1	9.495	.027
	Reformulations	3.573	.002	G-CI vs. G-NHL	T3	1.015	1	12.896	.004
				T1 vs. T2	G-CI	–0.614	1	9.297	.030
				T1 vs. T3		–1.092	1	29.399	<.001
				T1 vs. T4		–0.611	1	9.213	.030

Note. df: degrees of freedom; χ²: chi-square; G-CI: children with CIs; G-NHA: children with NH of the same chronological age; G-NHL: children with NH with similar language production abilities; NH: normal hearing.

Figure 1.

(a) Child Initiatives and (b) Adequate Responses Adjusted Means by Group and Session.

Aim 2: maternal input between groups and sessions

Analysis showed only main effects of Group, F(2, 48) = 19.137, p < .001, and Session, F(3, 147) = 7.793, p < .001, for mothers’ directives and significant Group × Session interactions for all remaining categories (see Table 2).

Separate follow-up pairwise comparisons for each language category showed that G-CI mothers produced more questions, labeling, comments, and reformulations, but less parallel talk, at T4 than at the previous sessions. Mothers generally reduced their overall use of directives over time (T3–T4: z = 3.105, p = .011); however, the overall proportions produced by G-CI mothers were significantly higher than those of G-NHL mothers (z = 4.884, p < .001) and marginally higher than those of G-NHA mothers (z = 2.312, p = .054). Tests also showed that G-CI mothers produced significantly different proportions of FLTs when compared with the mothers of both NH groups: at T1, less parallel talk than G-NHA mothers, but more labeling and parallel talk than G-NHL mothers; at T2 and T3, more labeling than G-NHA mothers; also at T3, fewer questions but more reformulations than G-NHL mothers; at T4, more labeling than mothers of both NH groups and more comments than the G-NHL group. See Table 2 for more details and Figure 2 for the adjusted mean proportions of each dependent variable across groups and sessions.

Figure 2.

(a) Maternal Questions, (b) Labeling, (c) Comments, (d) Parallel Talk, (e) Reformulations, and (f) Directives Adjusted Means by Group and Session.

Aim 3: the effect of maternal input on child language

Tables 3 and 4 report all the significant effects and/or interactions relative to the added predictor in the models and their relative significant pairwise comparisons, respectively.

Table 3.

Significant main effect or highest level interaction in the models including the other speaker’s characteristics as predictors.

Category	Added predictor	Main effect and/or interaction	F	df	p
Child	Mother
Initiatives	Questions	Questions × Group × Session	2.670	6, 149.75	.017
Initiatives	Reformulations	Reformulations	12.940	1, 146.150	<.001
Adequate responses	Questions	Questions × Session	4.244	3, 143.699	.007
	Labeling	Labeling × Group	5.728	2, 181.036	.004
	Parallel Talk	Parallel talk × Group	3.145	2, 174.781	.046
	Reformulations	Reformulations × Group	3.196	2, 136.702	.044
	Reformulations	Reformulations × Session	2.856	3, 156.723	.039
Mother	Child
Questions	Adequate responses	Adequate responses	8.116	1, 180.197	.005
Labeling	Adequate responses	Adequate responses × Group × Session	3.718	6, 154.557	.002
Reformulations	Initiatives	Initiatives	12.318	1, 170.117	<.001
Reformulations	Adequate responses	Adequate responses	50.814	1, 173.972	<.001

Table 4.

Significant pairwise comparisons relative to the G-CI group and significant slopes in child models when adjusted for maternal factors.

Dependent variable	Factor for which the slope is adjusted	Contrast	Value	df	χ²	p	Group or Session with significant slopes	Value	df	χ²	p
Child	Mother
Initiatives	Questions	G-CI vs. G-NHA at T4	–1.133	1	16.762	<.001	G-CI at T4	–0.850	1	16.067	<.001
Initiatives	Questions	G-CI group: T2 vs. T4	1.158	1	8.961	.050	G-CI at T4	–0.850	1	16.067	<.001
Adequate responses	Labeling	G-CI vs. G-NHA	0.485	1	10.919	.003	G-NHA	–0.495	1	17.616	<.001
Mother	Child
Labeling	Adequate responses	G-CI group: T3 vs. T4	1.433	1	15.124	.002	G-CI at T4	–1.201	1	11.949	<.001

Note. df: degrees of freedom; χ²: chi-square; G-CI: children with CIs; G-NHA: children with NH of the same chronological age; NH: normal hearing.

Initiatives

Model fit improved only when mothers’ questions, χ²₍₁₂₎ = 28.509, p = .005, and reformulations, χ²₍₁₂₎ = 30.922, p = .002, were added as predictors. In the model including Mothers’ questions, we found significant interactions with Group and Session. Pairwise comparisons showed significant differences between the G-CI and G-NHA groups at T4 and between T2 and T4 for the G-CI group. In all cases, only the (negative) slope for the G-CI group at T4 was significantly different from zero – see Figure 3(a). The model including mothers’ reformulations only showed a main effect of the added predictor (see Table 3), with children’s initiatives increasing as their mothers’ reformulations increased – see Figure 3(b).

Figure 3.

Plots Showing the (a) Effect of the Interaction Between Mothers’ Questions, Group, and Session and the (b) Effect of Reformulations on Children’s Initiatives in Interaction, and the (c) Effect of the Interaction Between Mothers’ Labeling and Group and the (d) Effect of Reformulations on Children’s Adequate Responses.

Adequate responses

Model fit improved when mothers’ questions, χ²₍₁₂₎ = 44.804, p < .001; labeling, χ²₍₁₂₎ = 26.951, p = .008; parallel talk, χ²₍₁₂₎ = 21.937, p = .038; reformulations, χ²₍₁₂₎ = 130.93, p < .001; or directives, χ²₍₁₂₎ = 22.151, p = .036, were added as predictors. The model including mothers’ labeling showed a significant interaction with Group: Pairwise comparisons showed significant differences between the G-CI and G-NHA groups, with only the G-NHA group showing a significant negative slope for labeling – see Figure 3(c). We also found a significant interaction between Group and reformulations and between Group and parallel talk; however, pairwise comparisons indicated no significant differences between groups. In the model including mothers’ reformulations as a predictor, there was only a general effect for which children’s adequate responses increased as a function of mothers’ reformulations, F(1, 141.806) = 86.544, p < .001; see Figure 3(d). Finally, the remaining models only showed significant interactions between the added predictor and Session (for maternal questions) or no significant interactions between the added predictor and Group and/or Session (for directives).

Aim 4: the effect of child language on maternal input

Tables 3 and 4 report all the significant effects and/or interactions relative to the added predictor in the models and their relative significant pairwise comparisons, respectively.

FLTs

Questions

Model fit improved with the addition of children’s adequate responses, χ²₍₁₂₎ = 31.324, p = .002, as predictor. However, the analysis only showed a significant main effect of adequate responses, F(1, 180.197) = 8.116, p = .005, with mothers’ questions increasing as a function of children’s responses – see Figure 4(a) – and no significant interactions between the added predictor and Group and/or Session.

Labeling

Model fit improved when adding child responses, χ²₍₁₂₎ = 36.114, p < .001. Analysis showed a significant interaction with Group and Session, with a significant difference between T3 and T4 for the G-CI group as regards responses (with a significant, negative slope at T4 only) – see Figure 4(b).

Comments, parallel talk, and directives

None of the child variables improved base LMMs on maternal comments, parallel talk, or directives.

Reformulations

Model fit improved when adding initiatives, χ²₍₁₂₎ = 41.041, p < .001, or responses, χ²₍₁₂₎ = 116.09, p < .001. However, in both models, there was only a main effect of the added predictor and no significant interactions. See Figure 4(c) and (d) for a visual depiction of these results.

Figure 4.

Plots Showing the (a) Effect of Children’s Adequate Responses on Mothers’ Questions, (b) Effect of the Interaction Between Children’s Adequate Responses, Group, and Session on Mothers’ Labeling, and Effects of (c) Children’s Initiatives and (d) Adequate Responses on Mothers’ Reformulations.

Discussion

The present study aimed to investigate the communicative characteristics of children with CIs and their mothers in interaction, whether and how they differ from those of mother–child dyads with NH, and whether mother and child influence each other over the first year after implantation. Results showed delayed communicative development for the children with CIs tested. Although all children increased the number of initiatives and adequate responses produced over time, the difference between the adequate responses produced by children with CIs and by their age-matched and language-matched peers became statistically significant 12 months after implantation. In other words, the trajectories of the adequate responses produced by children with CIs and by their peers with NH only coincided for the first 6 months observed; at 12 months post-implantation, children with CIs produced significantly fewer adequate responses than both NH groups. As shown in Figure 1, this is probably due to the fact that, while the other children significantly increased their production of adequate responses in the last 6 months observed, the number of adequate responses produced by children with CIs remained quite constant between 6 and 12 months after implantation. However, it should also be noted that the similar proportions of initiatives found at the last session across groups suggest that children with CIs may be on the right track to catch up with their peers. This idea is also reinforced by Lavelli et al.’s (2018) findings of comparable initiatives and adequate responses in Italian-speaking children with CIs 1 year later, at 2 years after implantation.

Concerning mothers’ use of FLTs, results showed significant differences between groups for labeling and comments. Mothers of children with CIs produced more labeling after implantation than those of age-matched children, in line with DesJardin et al. (2014). These differences may also depend on the fact that, although labeling was used by all mothers initially, only those of hearing children reduced its use when their child was about 20 months. The continuous and increasing use of these strategies by mothers of children with CIs throughout the whole observation period may indicate an attempt to facilitate language acquisition. The same concept may explain why mothers of children with CIs produced more comments than those of language-matched children with NH 1 year after implantation. In contrast, no differences were found post-implantation for parallel talk and reformulations: All mothers decreased their production of parallel talk and increased their production of reformulations over time.

Although there was a general tendency to reduce directives use with time, the proportions produced by G-CI mothers remained higher overall than those produced by mothers of language-matched children with NH. These results indicate that, at a comparable child language level, the mothers of children with CIs were more directive in interaction with their child, arguably in an attempt to redirect the child’s attention. Interestingly, preliminary analyses not including maternal education as a covariate had shown significantly higher proportions than age-matched children’s mothers too. Previous studies showing that the speech of mothers of children with CIs is associated with more directiveness than that of mothers of age-matched children with NH have not always taken into account the effect of maternal education or explicitly reported whether groups were socioeconomically comparable (e.g. Adi-Bensaid & Greenstein, 2020; Fagan et al., 2014; cf. Dirks et al., 2020). The finding that, in our study, the difference in directives use between mothers of age-matched children with or without NH becomes marginal when maternal education is accounted for reinforces the idea that maternal education affects maternal input characteristics (Hoff, 2006) and suggests that the mixed evidence found in the literature may also partly depend on whether or not studies controlled for differences in maternal education. Nevertheless, our study still shows a tendency for mothers of children with CIs to produce more directives (see also Figure 2). One explanation for this result seemingly in contrast with Dirks et al.’s (2020) is related to the degree of children’s deafness (moderate there vs severe to profound here), with mothers tending to be more directive only when their child’s deafness is more severe.

Importantly, we found that mothers’ communicative behavior affected their children’s communication. Independently of hearing status, children produced more initiatives and adequate responses with more maternal reformulations. However, not all techniques were beneficial: More maternal questions the to effect to reduce the opportunities to initiate a turn of the children with CIs. Interestingly, the performance of children with CIs was not affected by maternal use of labeling: we found negative effects on responses only in age-matched NH children. Taken together, these important results indicate that more repetitions and expansions are effective strategies to support children’s interaction behaviors, but that asking too many questions may deny children with CIs the opportunity to take initiative in interaction and thus hinder their language and communicative development. Note that this negative effect may be due to the presence of many closed-ended questions, which limit the linguistic material that children can produce in response but which were not distinguished from open-ended questions in our study. We also found that communicative strategies such as labeling yield negative effects on hearing children at later stages of development (i.e. in the age-matched group, not in the 6 months younger, language-matched group), which suggests that they are detrimental once the child has reached a certain language proficiency level.

Effects of the interlocutor’s communicative behavior were also found in the other direction. Mothers generally produced more questions when their children produced more adequate responses; they also increased their use of reformulations when their children produced more initiatives and adequate responses. Relative to the group of children with CIs, we found a negative effect of children’s adequate responses on maternal labeling use 1 year after implantation. This suggests that mothers of children with CIs use more labeling to support language acquisition and only reduce its proportions when their child’s perceived language quality starts improving.

For all the other FLTs, we did not find an effect on child’s communicative behavior. Because we found that mothers of children with CIs produced significantly more comments 1 year after implantation and significantly more directives in general, this important finding implies that mothers of DHH children do not always promptly adjust their communicative behavior to the child’s abilities. Although all mothers tend to take more of a lead in interaction with their children in the first few months of their life, those of DHH children may continue to be overly dominant, irrespectively of their child’s communicative improvements, possibly in the fear that their child’s language skills are not advancing enough.

In sum, the present study showed communicative differences between groups of DHH and NH children and their mothers but also important bidirectional influences between mother and child in interaction. The spoken communicative development of children with CIs appeared to be delayed compared with that of NH peers and to be negatively influenced by mothers’ use of questions. Mothers’ production of FLTs was affected by the child’s ability to hear (e.g. in the use of directives), by the child’s communicative behavior and perceived language and communicative skills (e.g. in the use of reformulations), or by both (e.g. in the use of labeling). These results may have important implications for clinical practice: On the grounds that teaching parents has proven successful in previous research (Kaiser et al., 1998), professionals may advise parents on the techniques to use to most successfully promote language acquisition in children with CIs and to remediate delays. The present study included small samples within groups, did not differentiate between open and closed questions, and did not investigate whether parental input was coordinated with children’s attention (important for successful language acquisition; Houston, 2022) and/or also delivered using nonverbal cues such as gestures or facial expressions. Further research will contribute to establish whether the findings of the present study hold with larger participant samples, more comprehensive measures, and/or different languages.

Footnotes

Appendix 1

Table 5.

Child communicative functions and maternal Facilitative Language Techniques of interest.

Technique	Description	Example
Child
Initiatives	Any expression (affirmation, comment, exclamation) to start a conversation.	Child asks the mother, “Can we play with the car?”
Adequate responses	Any vocal response semantically and/or linguistically related to a maternal question, or any partial or total repetition of the maternal enunciation.	Mother asks “Are you cooking peas?”, Child replies “Yes” or “Yes, I cook peas.”
Mother
Questions	Phrases or sentences addressed to the child to which the child can respond with a simple or complex statement.	Mother asks, “Do you want to play with the farm?”
Labels	Phrases or sentences that state object names or types; child may or may not look directly at the object.	Child plays with a toy dog, mother says, “This is a dog.”
Comments	Phrases or sentences to maintain or promote the communicative exchange, to give positive reinforcement or feedback.	Child yawns, mother says, “Oh, you look so tired.”
Parallel Talk	Phrases or sentences that provide linguistic input directly on what is happening while the child is paying attention to the activity.	Child pretends to drink, mother says, “You’re drinking.”
Reformulations	Direct repetition of the child’s previous production, or sentence resuming the child’s previous statement by inserting one or more morphemes or words or correcting the production, whether or not new information is added.	Child says, “dog drink”, mother says, “the dog is drinking”
Directives	Statements used to tell the child to do something.	Mother says, “Take that toy.”

Acknowledgements

We would like to thank Tamara Bastianello for her assistance with preliminary data analysis and all the families who participated in the study.

Author contribution(s)

Valentina Persici: Data curation; Formal analysis; Investigation; Visualization; Writing – original draft; Writing – review & editing.

Marika Morelli: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing – review & editing.

Manuela Lavelli: Conceptualization; Methodology; Supervision; Writing – review & editing.

Elena Florit: Conceptualization; Supervision; Writing – review & editing.

Letizia Guerzoni: Conceptualization; Data curation; Supervision; Writing – review & editing.

Domenico Cuda: Conceptualization; Supervision; Writing – review & editing.

Christine Yoshinaga-Itano: Conceptualization; Writing – review & editing.

Marinella Majorano: Conceptualization; Methodology; Writing – original draft; Writing – review & editing.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of Verona and Cochlear S.r.l. (Joint-Project Grant 2016).

ORCID iD

Valentina Persici

References

Adi-Bensaid

Greenstein

(2020). The effect of hearing loss on the use of lexical categories by Hebrew-speaking mothers of deaf children with cochlear implants. International Journal of Pediatric Otorhinolaryngology, 131, Article 109880. https://doi.org/10.1016/j.ijporl.2020.109880

Akhtar

Dunham

P. J.

(1991). Directive interactions and early vocabulary development: The role of joint attentional focus. Journal of Child Language, 18(1), 41–49. https://doi.org/10.1017/S0305000900013283

Ambrose

S. E.

Walker

E. A.

Unflat-Berry

L. M.

Oleson

J. J.

Moeller

M. P.

(2015). Quantity and quality of caregivers’ linguistic input to 18-month and 3-year-old children who are hard of hearing. Ear Hear, 36(1), 48–59. https://doi.org/10.1097/AUD.0000000000000209

Bates

Mächler

Bolker

Walker

(2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1), 1–48. https://doi.org/10.18637/jss.v067.i01

Bergeson

T. R.

Miller

R. J.

McCune

(2006). Mothers’ speech to hearing-impaired infants and children with cochlear implants. Infancy, 10(3), 221–240. https://doi.org/10.1207/s15327078in1003_2

Caselli

M. C.

Bello

Rinaldi

Stefanini

Pasqualetti

(2015). Il Primo Vocabolario del Bambino: Gesti, Parole e Frasi. Valori di riferimento fra 8 e 36 mesi delle Forme complete e delle Forme brevi del questionario MacArthur-Bates CDI [The First Vocabulary of the Child: Gestures, Words and Sentences. Reference Values between 8 and 36 months of the Complete and Short Forms of the MacArthur-Bates CDI questionnaire]. Franco Angeli.

Caselli

M. C.

Rinaldi

Varuzza

Giuliani

Burdo

(2012). Cochlear implant in the second year of life: Lexical and grammatical outcomes. Journal of Speech, Language, and Hearing Research, 55, 382–394. https://doi.org/10.1044/1092-4388(2011/10-0248)382

Clark

J. G.

(1981). Uses and abuses of hearing loss classification. ASHA: A Journal of the American Speech-Language-Hearing Association, 23(7), 493–500.

Cruz

Quittner

A. L.

Marker

DesJardin

J. L.

, & the CDaCI Investigative Team. (2013). Identification of effective strategies to promote language in deaf children with cochlear implants the CDaCI Investigative Team. Child Development, 84(2), 543–559. https://doi.org/10.1111/j.1467-8624.2012.01863.x

10.

Cupples

Ching

T. Y. C.

Button

Seeto

Zhang

Whitfield

Gunnourie

Martin

Marnane

(2018). Spoken language and everyday functioning in 5-year-old children using hearing aids or cochlear implants. International Journal of Audiology, 57(Suppl. 2), S55–S69. https://doi.og/10.1080/14992027.2017.1370140

11.

Curtin

Dirks

Cruice

Herman

Newman

Rodgers

Morgan

(2021). Assessing parent behaviours in parent-child interactions with deaf and hard of hearing infants aged 0-3 years: A systematic review. Journal of Clinical Medicine, 10, Article 3345. https://doi.org/10.3390/jcm10153345

12.

De Rosario-Martinez

(2015). phia: Post-hoc interaction analysis (R package version 0.2-1). https://cran.r-project.org/package=phia

13.

DesJardin

J. L.

Doll

E. R.

Stika

C. J.

Eisenberg

L. S.

Johnson

K. J.

Ganguly

D. H.

. . .Henning

S. C.

(2014). Parental support for language development during joint book reading for young children with hearing loss. Communication Disorders Quarterly, 35(3), 167–181. https://doi.org/10.1177/1525740113518062

14.

DesJardin

J. L.

Eisenberg

L. S.

(2007). Maternal contributions: Supporting language development in young children with cochlear implants. Ear and Hearing, 28(4), 456–469. https://doi.org/10.1097/AUD.0b013e31806dc1ab

15.

Dirks

Stevens

Kok

Frijns

Rieffe

(2020). Talk with me! Parental linguistic input to toddlers with moderate hearing loss. Journal of Child Language, 47(1), 186–204. https://doi.org/10.1017/S0305000919000667

16.

Fagan

M. K.

Bergeson

T. R.

Morris

K. J.

(2014). Synchrony, complexity and directiveness in mothers’ interactions with infants pre- and post-cochlear implantation. Infant Behavior and Development, 37, 249–257. https://doi.org/10.1016/j.infbeh.2014.04.001

17.

Farran

L. K.

Lederberg

A. R.

Jackson

L. A.

(2009). Maternal input and lexical development: The case of deaf pre-schoolers. International Journal of Language and Communication Disorders, 44(2), 145–163. https://doi.org/10.1080/13682820801973404

18.

Flipsen

Parker

R. G.

(2008). Phonological patterns in the conversational speech of children with cochlear implants. Journal of Communication Disorders, 41(4), 337–357. https://doi.org/10.1016/j.jcomdis.2008.01.003

19.

Fulcher

Purcell

A. A.

Baker

Munro

(2012). Listen up: Children with early identified hearing loss achieve age-appropriate speech/language outcomes by 3years-of-age. International Journal of Pediatric Otorhinolaryngology, 76(12), 1785–1794. https://doi.org/10.1016/j.ijporl.2012.09.001

20.

Geers

A. E.

Moog

J. S.

Biedenstein

Brenner

Hayes

(2009). Spoken language scores of children using cochlear implants compared to hearing age-mates at school entry. Journal of Deaf Studies and Deaf Education, 14(3), 371–385. https://doi.org/10.1093/deafed/enn046

21.

Guasti

M. T.

Papagno

Vernice

Cecchetto

Giuliani

Burdo

(2014). The effect of language structure on linguistic strengths and weaknesses in children with cochlear implants: Evidence from Italian. Applied Psycholinguistics, 35(4), 739–764. https://doi.org/10.1017/S0142716412000562

22.

Hoff

(2006). How social contexts support and shape language development. Developmental Review, 26, 55–88. https://doi.org/10.1016/j.dr.2005.11.002

23.

Holzinger

Dall

Sanduvete-Chaves

Saldanã

Chacón-Moscoso

Fellinger

(2020). The impact of family environment on language development of children with cochlear implants: A systematic review and meta-analysis. Ear and Hearing, 41(5), 1077–1091. https://doi.org/10.1097/AUD.0000000000000852

24.

Hothorn

Bretz

Westfall

(2008). Simultaneous inference in general parametric models. Biometrical Journal, 50(3), 346–363.

25.

Houston

D. M

. (2022). A framework for understanding the relation between spoken language input and outcomes for children with cochlear implants. Child Development Perspectives, 00, 1–7. https://doi.org/10.1111/cdep.12443

26.

Justice

L. M.

Weber

S. E.

Ezell

H. K.

Bakeman

(2002). A sequential analysis of children’s responsiveness to parental print references during shared book-reading interactions. American Journal of Speech-Language Pathology, 11(1), 30–40. https://doi.org/10.1044/1058-0360(2002/004)

27.

Kaiser

A. P.

Hancock

T. B.

Hester

P. P.

(1998). Parents as cointerventionists: Research on applications of naturalistic language teaching procedures. Infants and Young Children, 10(4), 46–55. https://doi.org/10.1097/00001163-199804000-00007

28.

Kondaurova

M. V.

Bergeson

T. R.

(2011). The effects of age and infant hearing status on maternal use of prosodic cues for clause boundaries in speech. Journal of Speech, Language, and Hearing Research, 54(3), 740–754. https://doi.org/10.1044/1092-4388(2010/09-0225)

29.

Kondaurova

M. V.

Fagan

M. K.

Zheng

(2020). Vocal imitation between mothers and their children with cochlear implants. Infancy, 25(6), 827–850. https://doi.org/10.1111/infa.12363

30.

Lavelli

Majorano

Guerzoni

Murri

Barachetti

Cuda

(2018). Communication dynamics between mothers and their children with cochlear implants: Effects of maternal support for language production. Journal of Communication Disorders, 73, 1–14. https://doi.org/10.1016/j.jcomdis.2018.03.001

31.

Zhao

Tao

Zhang

Yao

Zheng

(2020). Trajectory of auditory and language development in the early stages of pre-lingual children post cochlear implantation: A longitudinal follow up study. International Journal of Pediatric Otorhinolaryngology, 128, Article 109720. https://doi.org/10.1016/j.ijporl.2019.109720

32.

Longobardi

(1992). Funzione comunicativa del comportamento materno e sviluppo comunicativo-linguistico del bambino nel secondo anno di vita [Communicative functions of maternal behavior and the communicative-linguistic development of children in the second year of life]. Giornale Italiano Di Psicologia, 3, 425–448.

33.

Luckner

J. L.

Cooke

(2010). A summary of the vocabulary research with students who are deaf or hard of hearing. American Annals of the Deaf, 155(1), 38–67. https://doi.org/10.1353/aad.0.0129

34.

Lund

(2016). Vocabulary knowledge of children with cochlear implants: A meta-analysis. Journal of Deaf Studies and Deaf Education, 21(2), 107–121. https://doi.org/10.1093/deafed/env060

35.

MacWhinney

(2000). The CHILDES Project: Tools for analyzing talk. Third Edition. Lawrence Erlbaum Associates.

36.

Majorano

Brondino

Guerzoni

Murri

Ferrari

Lavelli

. . . Persici

(2021). Do acoustic environment characteristics affect the lexical development of children with cochlear implants? A longitudinal study before and after cochlear implant activation. American Journal of Audiology, 30, 602–615. https://doi.org/10.1044/2021_AJA-20-00104

37.

Majorano

Brondino

Morelli

Ferrari

Lavelli

Guerzoni

. . . Persici

(2020). Preverbal production and early lexical development in children with cochlear implants: A longitudinal study following pre-implanted children until 12 months after cochlear implant activation. Frontiers in Psychology, 11, Article 3164. https://doi.org/10.3389/fpsyg.2020.591584

38.

Majorano

Guidotti

Guerzoni

Murri

Morelli

Cuda

Lavelli

(2018). Spontaneous language production of Italian children with cochlear implants and their mothers in two interactive contexts. International Journal of Language and Communication Disorders, 53(1), 70–84. https://doi.org/10.1111/1460-6984.12327

39.

Moneglia

Cresti

(1997). Intonazione e criteri di trascrizione del parlato [Intonation and transcription criteria for speech]. In Bortolini

Pizzuto

(Eds.), Il progetto CHILDES (pp. 57–90). Del Cerro.

40.

Nelson

(1973). Structure and strategy in learning to talk. Monographs of the Society for Research in Child Development, 38(1/2), Article 1. https://doi.org/10.2307/1165788

41.

Nittrouer

Muir

Tietgens

Moberly

A. C.

Lowenstein

J. H.

(2018). Development of phonological, lexical, and syntactic abilities in children with cochlear implants across the elementary grades. Journal of Speech, Language, and Hearing Research, 61(10), Article 2561. https://doi.org/10.1044/2018_JSLHR-H-18-0047

42.

Olswang

Stoel-Gammon

Coggins

Carpenter

(1987). Assessing Linguistic Behaviour (ALB). University of Washington Press.

43.

Quittner

A. L.

Leibach

Marciel

(2004). The impact of cochlear implants on young deaf children: New methods to assess cognitive and behavioral development. Archives of Otolaryngology – Head and Neck Surgery, 130, 547–554. https://doi.org/10.1001/archotol.130.5.547

44.

R Development Core Team. (2016). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.r-project.org/

45.

Rinaldi

Baruffaldi

Burdo

Caselli

M. C.

(2013). Linguistic and pragmatic skills in toddlers with cochlear implant. International Journal of Language and Communication Disorders, 48(6), 715–725. https://doi.org/10.1111/1460-6984.12046

46.

Rowe

M. L.

Leech

K. A.

Cabrera

(2017). Going beyond input quantity: Wh-questions matter for toddlers’ language and cognitive development. Cognitive Science, 41, 162–179. https://doi.org/10.1111/cogs.12349

47.

Smith

N. A.

McMurray

(2018). Temporal responsiveness in mother–child dialogue: A longitudinal analysis of children with normal hearing and hearing loss. Infancy, 23(3), 410–431. https://doi.org/10.1111/infa.12227

48.

Spencer

P. E.

Marschark

(2010). Evidence-based practice in educating deaf and hard of hearing students. Oxford University Press.

49.

Szagun

Stumper

(2012). Age or experience? The influence of age at implantation and social and linguistic environment on language development in children with cochlear implants. Journal of Speech, Language, and Hearing Research, 55(6), 1640–1654. https://doi.org/10.1044/1092-4388(2012/11-0119)

50.

Szarkowski

Young

Matthews

Meinzen-Derr

(2020). Pragmatics development in deaf and hard of hearing children: A call to action. Pediatrics, 146(Suppl. 3), S310–S315. https://doi.org/10.1542/peds.2020-0242L

51.

Tompkins

Zucker

T. A.

Justice

L. M.

Binici

(2013). Inferential talk during teacher-child interactions in small-group play. Early Childhood Research Quarterly, 28(2), 424–436. https://doi.org/10.1016/j.ecresq.2012.11.001

52.

Uhler

Yoshinaga-Itano

Gabbard

S. A.

Rothpletz

A. M.

Jenkins

(2011). Longitudinal infant speech perception in young cochlear implant users. Journal of the American Academy of Audiology, 22(3), 129–142. https://doi.org/10.3766/jaaa.22.3.2

53.

Vanormelingen

De Maeyer

Gillis

(2015). Interaction patterns of mothers of children with different degrees of hearing: Normally hearing children and congenitally hearing-impaired children with a cochlear implant. International Journal of Pediatric Otorhinolaryngology, 79(4), 520–526. https://doi.org/10.1016/j.ijporl.2015.01.020

54.

Vihman

M. M.

McCune

(1994). When is a word a word? Journal of Child Language, 21(3), 517–542. https://doi.org/10.1017/S0305000900009442

55.

Wang

Jung

Bergeson

T. R.

Houston

D. M.

(2020). Lexical repetition properties of caregiver speech and language development in children with cochlear implants. Journal of Speech, Language, and Hearing Research, 63(3), 872–884. https://doi.org/10.1044/2019_JSLHR-19-00227

56.

Weizman

Z. O.

Snow

C. E.

(2001). Lexical input as related to children’s vocabulary acquisition: Effects of sophisticated exposure and support for meaning. Developmental Psychology, 37(2), 265–279. https://doi.org/10.1037/0012-1649.37.2.265

57.

Wie

O. B.

Torkildsen

von

Schauber

Busch

Litovsky

(2020). Long-term language development in children with early simultaneous bilateral cochlear implants. Ear and Hearing, 41, 1294–1305. https://doi.org/10.1097/aud.0000000000000851

58.

Yoder

P. J.

Warren

S. F.

Watson

A. L.

(2001). Important distinctions in measuring maternal responses to communication in prelinguistic children with disabilities. Communication Disorders Quarterly, 22(3), 135–147.

59.

Yoshinaga-Itano

Sedey

A. L.

Wiggin

Chung

(2017). Early hearing detection and vocabulary of children with hearing loss. Pediatrics, 140(2), Article 20162964. https://doi.org/10.1542/peds.2016-2964