Despite early identification and advancements in cochlear implant and hearing aid technology, delays in language skills in deaf children continue to exist. Good-quality parent–child interaction (PCI) is a key predictor for the successful development of deaf children’s signed and/or spoken language. As good-quality PCI is widely acknowledged to be significant for child language development in deaf children, then we must have the necessary tools to assess it. This review answers three questions: Which parent behaviours are assessed in PCI? How are they assessed? And which parent interaction behaviours are positively correlated with deaf children's language skills?
Despite earlier identification and advancements in hearing aid and cochlear implant technology, delays in receptive and expressive language skills in deaf children continue to exist [9,10][7][8]. Many studies have found the quantity and quality of parental interaction to be one of the main predictors of deaf children’s language outcomes [11,12,13,14,15][9][10][11][12][13]. Parents who have not yet developed skills in effectively communicating with their deaf child may provide lower-quality language input [16][14], which in turn affects the child’s language development.
To illustrate, studies have reported that hearing parents of deaf children can often be more directive in their interactions compared to deaf parents of deaf children and hearing parents of hearing children [9,17][7][15]. This manifests as increased interruptions to the child’s attention by parents initiating new, unrelated activities [18][16]. Hearing parents also elicit language from their deaf child through requests rather than conversations, meaning deaf children have less experience of two-way interaction and receive less feedback on their communicative attempts [19][17]. An important foundation for language development is joint attention, i.e., two people with a mutual focus. Hearing parents struggle to establish and maintain this behaviour with their young deaf infants [20,21][18][19]. Higher rates of directive behaviours from hearing parents of deaf children [17[15][20],22], are less conducive for maintaining attention. The mismatch of hearing status means that hearing parents need to adapt their communication skills to attain successful joint engagement in by gaining or waiting for the deaf child’s attention before starting to communicate and sequentially shifting attention between the environment/objects and each other. In comparison, deaf parents are using these social engagement strategies at an early age with their deaf infants [23,24][21][22] and we see an earlier tuning in of the deaf child’s gaze [25][23].
Paper No |
First Author | Year | Reported Country of Study | Study Design | Degree of Hearing Loss | No of Dyads | PCI Behaviours Assessed | PCI Measure (Method) | Child Lang Assessed? | Risk of Bias |
---|---|---|---|---|---|---|---|---|---|---|
1 | Beatrijs. W., et al. [21] | 2019 | Belgium | Two between-groups, observational studies |
No Report | 13 | Attention-Getting Strategies |
Coding | N | Moderate |
2 | DesJardin, J. L. [27] | 2006 | USA | Within-group, observational study |
Mod–Prof | 32 | Attention-Getting Strategies and Parental Communication | Coding | Y | Moderate |
3 | Loots, G. et al. [28] | 2003 | Belgium | Between-groups, observational study |
Mod–Prof | 33 | Attention-Getting Strategies |
Coding | N | Low |
4 | Waxman, R. et al. [29] | 1997 | USA | Between-groups, observational study |
Mild–Prof | 77 | Attention-Getting Strategies |
Coding | N | Moderate |
5 | Chasin, J. et al. [30] | 2008 | UK | Between-groups, observational study |
Profound | 9 | Attention-Getting Strategies and Child Eye Gaze | Coding | N | Moderate |
6 | Harris, M. et al. [31] | 1989 | UK | Within-group, longitudinal observational case series |
Profound | 4 | Attention-Getting Strategies and Child Eye Gaze |
Coding | Y | Serious |
7 | Harris, M. et al. [32] | 1997 | Australia and UK |
Between-groups, observational study |
Profound | 11 | Attention-Getting Strategies and Child Eye Gaze |
Coding | N | Critical |
8 | Harris, M. et al. [32] | 2005 | UK | Between-groups, observational study |
Profound | 26 | Attention-Getting Strategies and Child Eye Gaze |
Coding | N | Moderate |
9 | Lederberg, A. R. et al. [33] | 1998 | USA | Between-groups, observational study |
Sev–Prof | 40 | Attention-Getting Strategies and Child Eye Gaze | Coding | Y | Moderate |
10 | Prendergast, S. G. et al. [34] | 1996 | USA | Between-groups, observational study |
Sev–Prof | 16 | Attention-Getting Strategies and Child Eye Gaze | Coding | N | Moderate |
11 | Gabouer, A. et al. [35] | 2018 | USA | Between-groups, observational study |
Sev–Prof | 18 | Attention-Getting Strategies and Joint Engagement | Coding | N | Serious |
12 | Loots, G. et al. [22] | 2005 | Belgium | Between-groups, observational study |
Mod–Prof | 31 | Attention-Getting Strategies and Joint Engagement | Coding | N | Low |
13 | Nowakowski, M. et al. [36] | 2009 | Canada | Between-groups, observational study |
Sev–Prof | 56 | Attention-Getting Strategies and Joint Engagement | Coding | Y | Moderate |
14 | Tasker, S. et al. [37] | 2010 | Canada | Between-groups, observational study |
Sev–Prof | 53 | Attention-Getting Strategies and Joint Engagement | Coding | Y | Low |
15 | Barker, D. H et al. [7] | 2009 | USA | Between-groups, observational study |
Sev–Prof | 185 | Joint Engagement | Coding | Y | Low |
16 | Cejas, I. et al. [8] | 2014 | USA | Between-groups, observational study |
Sev–Prof | 276 | Joint Engagement | Coding | Y | Moderate |
17 | Roos, C. et al. [38] | 2016 | Sweden | Within-group, observational study |
Sev–Prof | 12 | Joint Engagement | Coding | N | Moderate |
18 | Spencer, P. E. [39] | 2000 | USA | Between-groups, observational study |
Mod–Prof | 80 | Joint Engagement | Coding | N | Serious |
19 | Dirks, E. et al. [18] | 2019 | The Netherlands | Between-groups, observational study |
Mod | 51 | Joint Engagement and Parental Sensitivity | Existing Scale + Coding | Y | Low |
20 | Gale, E. et al. [40] | 2009 | USA | Between-groups, observational study |
Sev–Prof | 15 | Joint Engagement and Parental Sensitivity | Coding | Y | Moderate |
21 | Janjua, F. et al. [41] | 2002 | UK | Within-group, observational study |
Sev–Prof | 13 | Joint Engagement and Parental Sensitivity | Coding | Y | Serious |
22 | Lederberg, A. R. et al. [42] | 1990 | USA | Between-groups, observational study |
Mild–Prof | 82 | Joint Engagement and Parental Sensitivity | Novel Scale + Coding | Y | Moderate |
23 | Meadow-Orlans, K. P. et al. [43] | 1993 | USA | Between-groups, observational study |
Mod–Prof | 80 | Joint Engagement and Parental Sensitivity | Novel Scale + Coding | N | Moderate |
24 | Meadow-Orlans, K. P. et al. [16] | 1996 | USA | Between-groups, observational study |
Mod–Prof | 80 | Joint Engagement and Parental Sensitivity | Novel Scale + Coding | N | Moderate |
25 | Abu Bakar, Z. et al. [44] | 2010 | Not reported | Between-groups, observational study |
Sev–Prof | 18 | Parental Sensitivity | Novel Scale | N | Serious |
26 | Meadow-Orlans, K. P. et al. [45] | 1995 | USA | Within-group, observational study |
Mild–Prof | 43 | Parental Sensitivity | Novel Scales | N | Moderate |
27 | Lam-Cassettari, C. et al. [46] | 2015 | UK | Between-groups, intervention study |
Mod–Prof | 14 | Parental Sensitivity | Existing Scale | N | Moderate |
28 | Meadow-Orlans, K. P. [47] | 1997 | USA | Between-groups, observational study |
Mod–Prof | 40 | Parental Sensitivity | Novel Scales | N | Moderate |
29 | Pressman, L. J. et al. [48] | 1998 | USA | Between-groups, observational study |
Mild–Prof | 42 | Parental Sensitivity | Existing Scale | Y | Moderate |
30 | Pressman, L. J. et al. [49] | 1999 | USA | Between-groups, observational study |
Mild–Prof | 24 | Parental Sensitivity | Existing Scale | Y | Low |
31 | Spencer, P.E. [50] | 1996 | USA | Between-groups, observational study |
Mod–Prof | 43 | Parental Sensitivity | Novel Scale | Y | Low |
32 | Vohr, B. et al. [51] | 2010 | USA | Between-groups, observational study |
Mild–Prof | 58 | Parental Sensitivity | Existing Scale | Y | Low |
33 | Waxman, R. et al. [52] | 1996 | USA | Between-groups, observational study |
Mod–Prof | 30 | Parental Sensitivity | Coding | N | Moderate |
34 | Ambrose, S. E. [53] | 2016 | USA | Between-groups, observational study |
Mild–Prof | 48 | Parental Sensitivity | Coding | Y | Low |
35 | Caissie, R. et al. [54] | 1993 | Not reported. | Between-groups, observational study |
Sev–Prof | 11 | Parental Sensitivity | Coding | Y | Serious |
36 | Eddy, J. R. [55] | 1997 | Australia | Between-groups, observational study |
Sev–Prof | 18 | Parental Sensitivity | Coding | Y | Serious |
37 | Glanemann, R. et al. [56] | 2013 | Germany | Between-groups, intervention study |
Mod–Prof | 29 | Parental Sensitivity | Coding | Y | Moderate |
38 | Wedell-Monnig, J.; et al. [57] | 1980 | USA | Between-groups, observational study |
Sev–Prof | 12 | Parental Sensitivity | Coding | N | Serious |
39 | MacTurk, R. H. et al. [58] | 1993 | USA | Between-groups, observational study |
Mod–Prof | 40 | Parental Sensitivity and Child Eye Gaze | Novel Scales | N | Serious |
40 | Choo, D. et al. [59] | 2016 | Australia | Within-group, observational study |
Sev–Prof | 12 | Parental Sensitivity and Parental Communication (Comm.) | Novel Scale | N | Moderate |
41 | James, D. et al. [60] | 2013 | UK | Within-group, intervention study |
Profound | 3 | Parental Sensitivity and Parental Comm. | Existing Scale + Coding | Y | Serious |
42 | Nicastri, M. et al. [11] | 2020 | Italy | Between-groups, intervention study |
Profound | Not reported: 22 parents of 14 children | Parental Sensitivity and Parental Comm. | Existing Scale | Y | Moderate |
43 | Preisler, G. M. [61] | 1995 | Sweden | Within-group, observational study |
No Report | 14 | Parental Sensitivity and Parental Comm. | Coding | N | Serious |
44 | Quittner, A. L. et al. [12] | 2013 | USA | Between-groups, intervention study |
Sev–Prof | 285 | Parental Sensitivity and Parental Comm. | Scales (×2 existing, ×1 novel) | Y | Low |
45 | Quittner, A. L. et al. [62] | 2016 | USA | Between-groups, observational study |
Profound | 285 | Parental Sensitivity and Parental Comm. | Scales (×1 existing/×1 novel) | Y | Low |
46 | Ahmad, A. et al. [63] | 2016 | Australia | Between-groups, observational study |
Mild–Prof | 16 | Parental Communication | Coding | N | Moderate |
47 | Brown, P. M. et al. [64] | 2004 | Australia | Between-groups, observational study |
Profound | 20 | Play and Parental Communication | Coding | Y | Moderate |
48 | Chen, D. [65] | 1996 | USA | Between-groups, observational study |
Mod–Prof | 12 | Parental Communication | Coding | Y | Serious |
49 | DeVilliers, J. et al. [66] | 1993 | USA | Within-group, observational study |
Profound | 2 | Parental Communication | Coding | N | Critical |
50 | Morelock, M. et al. [67] | 2003 | USA/Australia | Between-groups, observational study |
Profound | 9 | Parental Communication | Coding | N | Serious |
51 | Roberts, M. [68] | 2019 | USA | Randomised controlled trial | Mod–Prof | 19 | Parental Communication | Coding | Y | Moderate |
52 | Koester, L. S. et al. [69] | 2010 | USA | Between-groups, observational study |
Mod–Prof | 61 | Parental Communication | Coding | N | Serious |
53 | Paradis, G. et al. [70] | 2015 | USA | Between-groups, observational study |
No Report | 60 | Touch and Parental Sensitivity | Existing Scale + Coding | N | Moderate |
54 | Pipp-Siegel, S. et al. [71] | 1998 | USA | Between-groups, observational study |
Mild–Prof | 48 | Touch and Parental Sensitivity | Existing Scale + Coding | N | Moderate |
55 | Abu-Zhaya, R. et al. [72] | 2019 | USA. | Between-groups, observational study |
Mild–Prof | 24 | Touch | Coding | N | Moderate |
56 | Gabouer, A. et al. [73] | 2020 | USA | Between-groups, intervention study |
Sev–Prof | 18 | Touch | Coding | N | Serious |
57 | Spencer, P.E. [74] | 1993a | USA | Between-groups, observational study |
Mod–Prof | 36 | Other: Maternal Comm. Modality | Coding | Y | Low |
58 | Spencer, P.E. [75] | 1993b | USA | Between-groups, observational study |
Mod–Prof | 7 | Other: Maternal Comm. Modality | Coding | Y | Moderate |
59 | Lederberg, A. R. et al. [76] | 2000 | USA | Between-groups, observational study |
Sev–Prof | 40 | Other: Maternal Comm. Modality | Coding | Y | Moderate |
60 | Depowski, N. et al. [77] | 2015 | USA | Between-groups, observational study |
Sev–Prof | 8 | Other: Type and Use of Gesture | Coding | N | Serious |
61 | Lieberman, A. et al. [78] | 2014 | USA | Between-groups, observational study |
Mod–Prof | 8 | Other: Maternal and Child Eye Gaze | Coding | Y | Moderate |
Most papers (66%, n = 40) used a coding system to assess PCI, often watching and coding films frame by frame, using software such as INTERACT (Mangold) and ELAN (Max Planck Institute). This method allowed an in-depth analysis of the behaviours focused on in RQ1. Thirteen (13) papers (22%) used Likert scales instead of coding and some scales were well-known and validated, while others were developed for the specific research study with little mention of pilot testing prior to their use. The Emotional Availability Scales [96][82] were used in 7 of the 11 papers that used validated scales. Nine papers (12%) used a combination of coding and scales. See Table 2.
Method of Assessing PCI |
---|
n | Papers (%) |
---|
Coding | 40 (66%) |
An existing, validated scale | 7 (12%) |
A novel scale | 4 (7%) |
A mix of validated and novel scales | 2 (3%) |
Coding and a validated scale | 4 (6%) |
Coding and a novel scale | 4 (6%) |
In the study from Nicastri et al. [13][11], parents received nine whole-group sessions and three individual sessions of training over 10.5 months. The intervention was based on the ‘It Takes Two to Talk’ Hanen program [100][83] and involved video modelling, where parents had opportunities to put their training into practice at home. Strategies within the program included waiting and observing the child, following the child’s lead, interpreting the child’s behaviour, parallel talk, and expanding and recasting the child’s language. The authors reported significant gains in parent communication behaviours and parental sensitivity post-intervention and noted that parents in the treatment group had children with significantly better language skills, when formally assessed three years post treatment.
A pilot RCT [34][68] involved parents receiving weekly, hour-long sessions for six months, where they were explicitly taught to use strategies to promote early communication. The authors referred to methods such as enhanced milieu teaching [101][84], prelinguistic milieu teaching [102][85], and The Hanen Program It Takes Two to Talk [100][83]. Examples of strategies include sitting face to face, using gestures, imitating/mirroring the child’s actions, and turn taking. The study reported that parents in the treatment group increased their use of communication support strategies by 17% compared to 2% in the control group., There was a large effect size of 1.09 (p = 0.03) for the difference in gains in deaf children’s prelinguistic speech skills between the treatment and control groups.
The four main areas of PCI uncovered have some parallels with the review on children with language difficulties by Roberts and Kaiser [30][86], where the three most measured parent strategies were: parent responsiveness, use of language models and rate of communication. Similarly, a review by Holzinger and colleagues [104][87] on children with cochlear implants uncovered family involvement and parental linguistic input as key themes in their results. Additionally, within PCI research in the hearing population, the same set of behaviours are commonly measured [35][88].
The most prominent way of assessing PCI was with coding systems to analyse interactions. However, coding methods differed depending on the authors’ research focus. Some of the coding systems referred to well-known frameworks such as those from Waxman and Spencer [44][29], where attention-getting behaviours are well described and the coding scheme from Adamson, Bakeman, and Deckner [93][79], which includes 11 states of joint engagement. Other coding systems were created for the purposes of the particular study and papers did not report on the piloting of coding prior to their use. Behavioural observation is the ideal method for assessing the quality of interactions and reduces the risk of bias that may arise from the use of self-reporting tools [107][89]. Lotzin et al. [35][88] also limited their review of PCI assessments to objective instruments, with all 24 of their included measures being validated rating scales. Longer periods of joint engagement, increased parental sensitivity and a range of facilitative language techniques were all correlated with higher levels of language in deaf children. Parents with higher rates of maternal sensitivity and language stimulation have a greater effect on their child’s expressive language scores over time [13,64][11][49]. An important correlation highlighted by Vohr et al. [66][51] was that parents with more support and higher SES had decreased intrusiveness, directiveness and negative regard. The better supported a parent is, the more sensitive, responsive, and positive they will be in their interactions. Hintermair [109][90] mirrored this finding in his study with parents of deaf children showing that child development profits from parents accessing ‘personal and social resources’ that influence their coping process and significantly lower stress. Furthermore, Zaidman-Zait et al. [110][91] found that higher levels of child acceptance were associated with lower levels of parenting stress in parents of deaf children.