International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 Contents lists available at ScienceDirect International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl Hear here: Children with hearing loss learn words by listening Joyce Lew a,b,*, Alison A. Purcell b, Maree Doble b, Lynne H. Lim a a National University Health System, National University of Singapore, Department of Otolaryngology, NUHS Tower Block, Level 7, 1E Kent Ridge Road, Singapore 119228, Singapore b The University of Sydney, Discipline of Speech Pathology, Faculty of Health Sciences, P.O. Box 170, Lidcombe, NSW 1825, Australia A R T I C L E I N F O A B S T R A C T Article history: Received 7 April 2014 Received in revised form 20 July 2014 Accepted 22 July 2014 Available online 12 August 2014 Objectives: Early use of hearing devices and family participation in auditory-verbal therapy has been associated with age-appropriate verbal communication outcomes for children with hearing loss. However, there continues to be great variability in outcomes across different oral intervention programmes and little consensus on how therapists should prioritise goals at each therapy session for positive clinical outcomes. This pilot intervention study aimed to determine whether therapy goals that concentrate on teaching preschool children with hearing loss how to distinguish between words in a structured listening programme is effective, and whether gains in speech perception skills impact on vocabulary and speech development without them having to be worked on directly in therapy. Method: A multiple baseline across subjects design was used in this within-subject controlled study. 3 children aged between 2:6 and 3:1 with moderate–severe to severe-profound hearing loss were recruited for a 6-week intervention programme. Each participant commenced at different stages of the 10-staged listening programme depending on their individual listening skills at recruitment. Speech development and vocabulary assessments were conducted before and after the training programme in addition to speech perception assessments and probes conducted throughout the intervention programme. Results: All participants made gains in speech perception skills as well as vocabulary and speech development. Speech perception skills acquired were noted to be maintained a week after intervention. In addition, all participants were able to generalise speech perception skills learnt to words that had not been used in the intervention programme. Conclusions: This pilot study found that therapy directed at listening alone is promising and that it may have positive impact on speech and vocabulary development without these goals having to be incorporated into a therapy programme. Although a larger study is necessary for more conclusive findings, the results from this preliminary study are promising in support of emphasise on listening skills within auditory-verbal therapy programmes. Crown Copyright ß 2014 Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Children Hearing loss Auditory-verbal Speech perception Vocabulary Speech 1. Introduction Hearing loss (HL) is the most common birth anomaly, affecting 1–3 of every 1000 newborns in developed societies [1–4]. Congenital hearing loss impacts language learning [5,6], speech development [5,7], psychosocial development [8–10], literacy [6,11,12], academic success [13] and employment outcomes * Corresponding author at: National University of Singapore, Department of Otolaryngology, NUHS Tower Block, Level 7, 1E Kent Ridge Road, Singapore 119228, Singapore. Tel.: +65 6772 5370; fax: +65 6775 3820. E-mail addresses: [email protected] (J. Lew), [email protected] (A.A. Purcell), [email protected] (M. Doble), [email protected] (L.H. Lim). [13–16]. As a result, hearing loss is identified as one of the most costly lifelong conditions [16]. The sequelae of lifelong events affected by hearing loss seem to begin in-utero [17,18]. Babies with typical hearing are born with skills that facilitate language development from the first moments of life [19,20]. At birth, infants have preferential attention to their own mother’s voices [21], human speech sounds over nonspeech sounds [19], as well as an ability to recognise their own native language spoken over other languages [22]. Newborns with typical hearing are also able to perceive fine acoustic differences between individual speech sounds [21,23,24]. Where children are born with hearing loss, their ability to access speech acoustic information for detecting and discriminating between speech sounds is diminished. A recent cortical study on the auditory pathways of children born with http://dx.doi.org/10.1016/j.ijporl.2014.07.029 0165-5876/Crown Copyright ß 2014 Published by Elsevier Ireland Ltd. All rights reserved. J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 hearing loss report correlation between increased length in auditory deprivation and decreased likelihood in achieving normal cortical responses to sound through the use of assistive hearing devices [25]. 1.1. The importance of early intervention The need to minimise the lifelong impact of permanent sensorineural hearing loss has resulted in a number of important changes to the management of young children with hearing loss. Increasingly widespread implementation of newborn hearing screening has enabled the early detection of hearing loss whilst advances in amplification devices have enabled most children born with hearing loss to have auditory access to speech sounds. Best practice for newborns with permanent hearing loss currently involves the detection of hearing loss by one month of age, amplification by 3 months and attendance at early intervention by 6 months [26–28]. This is frequently referred to as the 1-3-6 guidelines. There is a spectrum of early intervention choices available to parents, from programmes that aim to teach children how to communicate manually to programmes that teach children how to communicate orally (refer to Schwartz [29] for a description of the main options). For parents who would like their child with hearing loss to learn to speak, an early intervention choice that has been gaining in popularity is Auditory-Verbal Therapy (AVT) [30]. With an emphasis on the use of appropriate hearing technology for auditory stimulation, the fundamental goal in AVT is for children to use hearing as the primary sensory modality in developing spoken language [31]. AVT focuses on creating listening experiences so that the child’s newly acquired auditory potential is optimally used as they learn to process verbal language and speak [32]. Children enrolled in AVT have been reported to be successful at achieving age-appropriate verbal communication outcomes [33–35]. A longitudinal outcome study of children enrolled in an AVT programme by Dornan et al. [33,36] has shown that language growth with each year of therapy was comparable with that of normal hearing peers matched for language age, gender and socioeconomic status. These findings were supported by a more recent longitudinal study that examined the language development of children who received intervention in accordance with 1-3-6 guidelines [35]. In this study, Fulcher et al. found that 90% of children born with hearing loss achieved age-appropriate speech and language outcomes by 3 years of age [35]. Outcome studies of AVT have been important in supporting the overall efficacy of AVT programmes in the current drive towards evidence-based practice. In order for such positive outcomes to be achieved across all AVT programmes, it is important to have evidence to support practitioners in the use of clinical techniques as well as prioritising goals that drive each therapy session. Rhoades [30] raised concerns regarding the lack of research evidence in supporting the process of AVT; and noted the need for data to answer critical questions such as which clinical strategies used in AVT programmes are more efficient than others. The current guiding principle for goal setting in AVT indicates that practitioners are to follow the developmental patterns of typical children covering the areas of audition, speech, language, cognition and communication [31]. With respect to audition, there are a number of published developmental hierarchies used by AV practitioners to develop and refine a child’s listening skills. Some examples include the ‘Listening Skills Scale for Auditory-Verbal Therapy (LSSAVT)’ [32], ‘Targets for Auditory/Linguistic Learning’ [17], the ‘St. Gabriel’s Curriculum’ [37] and the ‘Listen Learn and Talk’ [38]. While these audition hierarchies are widely used and central to therapy planning, no known outcome study has been published on the efficacy of designing a therapy programme based on these hierarchies. 1717 Speech perception is a significant component of many widely used audition hierarchies, yet little is known regarding the efficacy of speech perception intervention with infants and preschool children with hearing loss. Research evidence is available only for the efficacy of speech perception intervention with adults and school-aged children. Although these intervention studies reported positive outcomes [39–42], their applicability to preschool children is limited because all of these programmes require participants to have prior linguistic and letter-to-sound knowledge in addition to being literate. Moreover, the programmes tracked the development of speech perception skills without measuring the development of other aspects of verbal communication skills such as vocabulary development and speech intelligibility, thereby failing to demonstrate any improvement in functional skills as a result of improved speech perception or auditory development. It is important to understand the essential role that speech perception intervention plays in AVT as teaching listening is one of its key elements. Several studies have found correlation between development of early speech perception and later vocabulary skills in young children with normal hearing [43–45]. In addition, prospective studies of typically developing infant speech perception and early language abilities report that infant speech perception skills accurately predict low vs normal language function up to 8 years of age [46–48]. These findings are exciting for children with hearing loss. If speech perception skills during infancy can predict language status in later childhood, perhaps early intervention of poor speech perception skills can intercept the course of delayed or impaired language development. In order to determine whether there is positive correlation between speech perception skills and vocabulary in pre-schoolers with hearing loss, Desjardin et al. [49] designed a speech perception test that is based on the repetition of nonsense syllables. This cross-sectional study found significant correlation between speech perception and vocabulary skills, thereby supporting the hypothesis that speech perception intervention may have a positive impact on the vocabulary of preschool children with hearing loss who are learning spoken language. An intervention study with school-aged children by Paatsch et al. [50] analysed the effects of speech production intervention as well as vocabulary intervention and found a causal impact on speech perception and reading development [50]. Although the impact of speech perception intervention on speech production and vocabulary skills cannot be ascertained from this study, it seems there is a relationship between development in speech production and/or vocabulary skills with development in speech perception in older children with hearing loss. Recent studies have found discontinuity between development in speech perception skills and early word learning for young preschool children with normal hearing [51–53]. These studies suggest that the processing demands of linking words to meaning are so great for the novice word-learner that they have difficulty in attending closely to the fine phonetic detail that is available in the speech signal. If children with normal hearing who are born ready for speech perception have difficulty attending to fine acoustic detail when they first begin learning words, what is the course of word learning in children with hearing loss who are still in the process of learning speech perception? A recent study on word learning processes in preschool children using cochlear implants who were enrolled in auditory-verbal programmes by Walker and McGregor [54] found that the children demonstrated delay in word learning despite early cochlear implantation. In comparison with typically developing peers, the children with hearing loss had difficulty with fast-mapping which is the word learning process of linking a word to its referents after only a few exposures [54]. Inferential fast-mapping whereby novel words were not directly marked by social or linguistic cues were particularly difficult for 1718 J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 below, which is aimed specifically at determining how consistent the participant is with word pairs that they had already achieved criterion in therapy. children with hearing loss, which may explain why children with hearing loss tend to learn fewer words incidentally than children with typical hearing [55]. Given the established link between speech perception and later language development in children with typical hearing, it is of interest to determine if children with hearing loss can learn words automatically in the process of learning better speech perception skills without vocabulary having to be incorporated into goals of auditory-verbal therapy. The answer to this question is important for professionals working with preschool children with hearing loss as it may allow the course of word learning to be sped up through more successful inferential learning. In addition to vocabulary and language development, intact speech perception skills are necessary for speech development, with hearing loss being a recognised causal factor in speech difficulties [56,57]. In spite of early identification and intervention for children born with hearing loss, a longitudinal study of speech development by Moeller et al. found that this group of children continues to have overall delayed speech development with particular difficulty in fricative and affricate production [7]. This study demonstrated that although early intervention has enormous benefits for children, listening and the development of skills associated with this sense continue to be difficult for children with hearing loss [7]. Noting the lack of intervention studies on the impact of speech perception intervention on speech production, Moeller [9] also suggested the implementation of intervention studies to determine whether focused listening practise could impact on speech development. Since there have been no known studies to support an assumption that word learning and speech development automatically follows improved speech perception skills, it is important to monitor whether they develop alongside speech perception skills for children with hearing loss. The aims of this study were therefore: (1) to determine whether speech perception intervention for preschool children with hearing loss is effective, (2) to determine whether development in speech perception skills has positive impact on vocabulary development, and (3) to determine whether development in speech perception has positive impact on speech development. 1.2. Speech Perception Education and Assessment Kit (SPEAK) 1.2.1. SPEAK-Intervention The intervention programme consisted of 11 levels of difficulty and commenced at the child’s performance level as determined by the SPEAK-Baseline. At each intervention session, the child worked through 2 sets of cards (A and B) with a parent as well as the therapist. There were 2 pairs of cards in each of sets A and B, as illustrated in Fig. 1 below. Intervention commenced with the first pair of words in set A. Each of the picture cards were hidden in a variety of objects (e.g., envelope, box, stuck to the bottom of a toy, etc.) and presented one at a time through audition in an auditory exposure phase. This phase involved the therapist taking an object that contained a picture card, saying the target word and having a peep at the picture in the object. The therapist would then turn to the child’s parent who will provide the child with another opportunity to listen to the target word without seeing the corresponding picture card in the same manner. The child is finally given an opportunity to open the object to see a picture card that represents the word. While the child looked at the picture, the therapist said 3 short phrases with each phrase comprising the target word. This process is repeated for the second picture card, yielding two target words for each child to practice perception with their parent. In the practice phase, both cards were first put in front of the parent within the child’s visual field, and the parent was asked to listen to the therapist say one of the words and point to the associated picture. This followed with the child having a turn at listening to a word and attempting to point at the corresponding picture. Both the parent and the child took turns practicing 5 times each before a new pair of cards was presented using novel objects and worked on in the same way. The 2 pairs of cards from Set A yielded a therapy score out of 10, which allowed the therapist to determine whether pass criterion had been reached. If the child was accurate for 8 or more times, Set B of the intervention session comprised of words from the next level of the intervention programme. Where the child did not meet the pass criterion with Set A of the therapy session, Set B comprised of a different set of words at the same level of difficulty as that of Set A. Feedback was given each time after the child and parent pointed to a card in the practice phase. When a mistake was made, the child was shown the correct picture while he heard a repetition of the target word. The parent and child were then be given a token reinforcement each time after feedback, In order to answer these questions, the Speech Perception Education and Assessment Kit (SPEAK) programme was developed. The work of Boothroyd [58] which depicts the spread of acoustic information available to listeners to distinguish between speech sounds was applied in the development of this 10-level intervention programme. The programme consisted of the (i) SPEAKIntervention, (ii) SPEAK-Baseline as well as (iii) SPEAK-Probes. The SPEAK programme commenced at level 1 with discriminating between long and short words that have differing numbers of syllables (e.g., ‘aeroplane’ vs ‘sun’), followed by single-syllabic words that have maximal differences in acoustic information available for the listener to discriminate between them (e.g., ‘bus’ vs ‘cloud’) in level 2. The latter stages of the programme consist of single-syllabic words that sound increasingly similar and are thus more difficult to discriminate between. The difficulty was increased by systematically decreasing the amount of acoustic information that was available to the listener to discriminate between each pair of words until level 10 which consists of words with minimal acoustic information between the word pairs. Although the level of speech perception difficulty are the same across sections of the SPEAK programme (Intervention, Baseline and Probes), word pairs used in each section of the SPEAK programme were not repeated in other sections of the programme. The exception to this is the consistency probe, further described [(Fig._1)TD$IG] Fig. 1. Illustration of each intervention session. J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 whether or not they picked correctly. At the end of each 30-min intervention session, the parent was given materials for home follow-up the next day. Follow-up consisted of a fresh set of cards at the highest level that the child had reached pass criteria and another set of cards at the next level at which the child had not achieved criterion. The intervention programme terminated after 18 sessions with the therapist or after the child achieved pass criterion at level 10 of the intervention programme, whichever occurred first. 1.2.2. SPEAK-Baseline The SPEAK-Baseline determined the level at which the child commenced the intervention programme as well as change in speech perception skills before and after the intervention programme. The baseline was constructed to match the 10 speech perception levels of the SPEAK-Intervention programme; the higher the level the finer the acoustic differences between the words. Each level of the SPEAK-Baseline consisted of four picture cards which were individually presented to the child as the therapist verbally labelled them without allowing the child to lip-read. The child then listened to a target word and pointed to the picture that he thought represented the target word for 10 trials at each level of the test. The SPEAK-Baseline yielded a score representing the number of times the child pointed correctly out of a maximum score of 100, as well as a skill level between 0 and 10. It was administered before the intervention programme to determine the level at which each child should commence the SPEAK-Intervention programme. The skill level was the highest level out of the 10 levels of the test at which the child scored at least 8 out of 10. The skill level is also the level at which participants commence SPEAK-Intervention. Token reinforcers were used to keep the participants engaged throughout the baseline measures. SPEAK-Baseline was re-administered after the SPEAK-Intervention phase to determine any changes in speech perception levels. 1.2.3. SPEAK-Probes Three progress probes were incorporated into the SPEAK programme in order to examine generalisation across theoretically related responses as well as to attain within-subject experimental control data. The first was the generalisation probe which consisted of 4 words represented by picture cards that were at the same level of difficulty as the child’s speech perception skill level. Words used in the generalisation probe were novel to words used in both the SPEAK-Intervention programme and the SPEAKBaseline. This probe was designed to provide information on how well the child generalised their speech perception skills from words used in the intervention programme to discriminating between untaught words. The second progress probe was the control probe which consisted of four novel words that were three levels above the child’s perceptual level. Participants’ performance on the control probe should ideally remain fairly constant throughout their participation in the intervention programme. The final probe, the consistency probe, comprised of 4 words at perceptual level that the child had been exposed to during the speech perception intervention programme. The consistency probe was designed to provide information on how consistent the child was with their speech perception skills with familiar words encountered during therapy at their skill level. With each probe, the child was presented with four picture cards in the same way as in the SPEAK-Baseline and then asked to point to a card labelled by the therapist for 10 trials. Review Board (Reference Number DSRB-D/11/016) and the University of Sydney Ethics Committee for research involving human subjects (Protocol Number 05-2011/13708). 2.1. Participants Participants were recruited from the AVT programme at the National University Hospital (Singapore) over a period of 12 months. The recruitment criteria required participants to be aged between 2:6 and 4:0, have bilateral permanent sensorineural hearing loss of at least moderate severity and be consistent users of hearing aids and/or cochlear implants. In addition, participants must score below the highest category score of four on the lowverbal version of the Early Speech Perception Test (ESP) [59]. Given the multilingual setting in Singapore, participants may be bilingual but use English as the main language of communication at home. All children who had been formally diagnosed with other impairments in addition to hearing loss were excluded from this study. From a total of 24 children aged between 2:6 and 4:0 attending the National University Hospital AVT programme, only 4 met these inclusion criteria. The families of 3 of these 4 children agreed to participate in the study. Of the 20 children who did not meet the inclusion criteria; one was excluded because of poor hearing aid compliance, 7 were excluded because English was not the main language spoken at home, 10 were excluded because they had multiple disabilities and 3 were excluded for achieving the highest category on the low-verbal version of the ESP. The details of each participant’s age at recruitment, severity of hearing loss as well as the type of listening devices used are summarised in Table 1. 2.2. Outcome measures A battery of pre- and post-intervention assessments was administered before and after the intervention programme was implemented, to evaluate the participants’ progress in speech perception, vocabulary and speech development. These tests are outlined below. 2.2.1. Speech perception The low-verbal version of the Early Speech Perception Test (ESP) [59] was used before and after SPEAK-Intervention to capture any changes in speech perception skills. In addition to the ESP being administered pre- and post-intervention, the SPEAK-Probes were administered throughout the study. The SPEAK-Baseline was administered to determine the starting level for the SPEAKIntervention. Both trained and untrained speech perception skills were monitored using three different progress probes to examine generalisation of skills while maintaining experimental control. Further information regarding the generalisation, control, and consistency probes can be found under SPEAK in the introduction. 2.2.2. Vocabulary Vocabulary was assessed using Versions A and B of the Peabody Picture Vocabulary Test – Fourth Edition (PPVT-4) [60] as well as the 2500+ Words List [61]. The PPVT-4 was selected to provide standard scores, percentile ranks and growth scale values (GSV). Table 1 Participant age, severity of hearing loss and listening devices used. Participant Age at recruitment Severity of hearing loss Listening devices S1 S2 S3 3:1 2:6 2:6 Severe–profound Moderate–severe Severe Bilateral cochlear implants Bilateral hearing aids Bilateral hearing aids 2. Method This study was conducted in accordance with ethical standards stipulated by the National Healthcare Group Domain Specific 1719 1720 J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 The GSV score was important as it provides evidence of change in vocabulary growth. This allowed any increase in vocabulary to be classified as either within expected developmental growth or exceeding typical developmental growth. The 2500+ Words List is a parent-administered receptive and expressive vocabulary checklist that requires parents to check off words that their child understands as well as uses spontaneously. If the parent identified that their child used a word that was not on the list, it was added to the checklist. This checklist also enabled monitoring of the participants’ vocabulary of English words as well as other languages that all participants were minimally exposed to in their respective environments. 2.2.3. Speech development The participants’ articulation skills were monitored using the Goldman–Fristoe Test of Articulation [62], percentage consonants correct (PCC) as well as a CASALA analysis for consonant phonological processes. The GFTA-2 measures the number of phonemes in the English Language the child is able to produce in comparison with a normative sample. Percentile rankings were calculated on the GFTA-2 and included for the purpose of giving a reference to compare each participant’s own progress. The GFTA-2 is scored on the basis of the child’s performance at targeted phonemes, with a single scored occurrence of each targeted phoneme in word initial, medial and final word positions where these phonemes occur in the English Language. For example, the phoneme/s/is scored at word initial position in ‘scissors’, word medial position in ‘pencils’ and word final position in ‘house’. The participants’ pronunciation of the/s/ phoneme is not scored in other occurrences in the speech sample, such as ‘this’ and ‘glasses’, which were included in the assessment for other targeted phonemes. In addition to percentile rankings, a more culturally sensitive percentage consonant correct (PCC) score was calculated. The PCC scores measure the percentage accuracy of the child’s phoneme productions without comparing it with a normative sample, thereby making each score directly comparable with the child’s previous scores on the same speech sample. PCC expresses the percentage of consonant sounds that were articulated correctly to provide a score of phonetic accuracy [63]. The PCC scores in this study were attained by comparing phonetic transcriptions of each participant’s speech productions from the GFTA-2 to that of a local Singaporean adult model. In order to analyse the development of participants’ speech error patterns, participants’ responses on the GFTA-2 were also analysed for phonological processes using the CASALA [64]. 2.3. Procedure This is a phase 1 study with a multiple baselines across participants design. Participants received 3 speech perception intervention sessions per week for 6 weeks or until they reached the ceiling of the intervention programme, whichever happened first. The baseline data collection period was randomly allocated to be either one or two weeks in duration (three or six sessions) as it is a feature of multiple-baseline across subjects experiments to extend the baseline phase for some participants. The withdrawal phase at the end of intervention programme took place over one week. The schedule of probes administered is illustrated in Fig. 2. The generalisation and control probes were administered in the baseline phase of the SPEAK programme and all three probes were administered in the intervention and withdrawal phases of the programme. During the baseline and withdrawal phases, probe data was collected 3 times a week. In the intervention phase, probe data was collected after therapy sessions 1, 3, 6, 9, 12, 15 and 18. 3. Results 3.1. SPEAK-Intervention All participants progressed through the SPEAK-Intervention programme. S1 and S2 advanced from level 3 to the highest level of the intervention programme at level 10; S1 reached criterion of the highest level of the intervention programme at the last scheduled intervention session whilst S2 completed the intervention programme after the 13th session. S3 started at level 1 and finished at level 7 at the end of the 18 therapy sessions (see Fig. 3). The SPEAK-Probes highlighted improvement on untrained words. The generalisation probes showed that the intervention impacted positively on untrained words at the same speech perception level that therapy was received for all participants. In contrast, the control data showed no generalisation to untrained words at increased levels of speech perception difficulty that had not been targeted in therapy (see Fig. 3). For S1, some of the control probe data were not collected because SPEAK-Intervention was initially designed with only 10 speech perception levels. S1’s improvement through the intervention programme was more rapid than anticipated resulting in the need for control probes beyond level 10. Consequently, control data for S1 were not available from the last probe point in the therapy phase and throughout the withdrawal phase. [(Fig._2)TD$IG] Fig. 2. Schedule of progress probes. [(Fig._3)TD$IG] J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 1721 Table 2 Speech discrimination results. S1 Pre S2 Pre Post Early Speech Perception (ESP) Test Total score (max = 36) 29 33 Category level (max = 4) 3 3 27 3 29 3 1 1 12 2 SPEAK-Baseline Results Total score (max = 100) Skill level (max = 10) 72 3 85 7 39 0 59 2 68 3 Post S3 89 7 Pre Post significant growth in receptive vocabulary, whilst the regression in S3’s vocabulary was not statistically significant. Using the 2500+ Words, receptive vocabulary growth was reported by the parents (21%, 161% and 210%, in subjects 1, 2, and 3, respectively). Positive gains were also reported in expressive vocabulary (18.6% and 215% and 80%, in subjects 1, 2, and 3, respectively). See Table 3. 3.3. Speech development Fig. 3. Speech perception probe results. 3.2. Pre and post outcome measures It was anticipated that T-tests would be carried out to compare measure change in the pre and post outcome measures. The residuals were screened for autocorrelation, and moderate to severe autocorrelation was found. No further statistical analysis was attempted, thus all data were analysed visually. 3.2.1. Speech perception Post-intervention, the total speech perception scores for all participants increased on the low-verbal version of the ESP (see Table 2). There was a single category level change for S3 while S1 and S2 maintained their category level close to ceiling of the test. 3.2.2. Vocabulary Percentile ranks on the PPVT-4 increased for S1 and S2, whilst a slight regression was noted for S3 (see Table 3). The PPVT-4 provides growth scale values (GSV) which determines whether changes in raw scores over time reach statistical significance. If the difference in GSV values between tests is greater than 8, statistically significant change may be deduced (p < 0.1) [60]. The GSV results indicate that S1 and S2 made statistically All participants showed an improvement in their speech development. S1’s scores remained within the normal range, however their percentile rank increased following intervention (see Table 4). S2’s speech development improved slightly but remained delayed and S3’s percentile range moved from the severely delayed range to a mild delay. There was also an increase in PCC scores for all participants following intervention. Participants made between 20% and 40% improvement in PCC accuracy (see Table 4). Results from the CASALA analysis of consonant phonological processes are presented in Table 5. A phonological process was deemed clinically relevant if it occurred at least 20% of the time. This was in accordance with McReynold and Elbert [65], who considered a phonological process to sufficiently affect a child’s speech to a degree that warrants remediation if it occurred at least 20% of the time. Phonological processes are mostly typical in a child’s speech, remediating spontaneously at different ages depending on the type of process. Some phonological processes do not occur frequently in typically developing speech and their frequent occurrence may indicate disordered speech development [66]. S1’s use of phonological processes did not reach clinical relevance throughout her participation. S2 had a slight reduction in the number of phonological processes after intervention. After intervention, S2 had a reduction in the use of atypical phonological processes from 3 to 2. While S3 increased the number of phonological processes affecting her speech from 2 to 3 processes, these processes were typical for their age. An increase in phonological processes for S3 is to be expected because there was an increase in the number of words used overall, from 19 to 53 words attempted in the GFTA-2. 3.4. Reliability 3.4.1. Treatment fidelity An independent Speech Language Pathologist reviewed 10% of the therapy sessions to determine whether the intervention protocol was adhered to. The reviewer marked the Speech Language Pathologist responses in whether she (i) presented the target words through audition before its visual representation, (ii) presented the target words 3 times in a phrase during speech perception intervention, (iii) took accurate therapy data, (iv) provided a token when the child was accurate and (v) taught on error responses and provided the child with a token. The J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 1722 Table 3 Vocabulary results. S1 S2 S3 Pre Post Pre Post Pre Post PPVT-4 Standard score Percentile rank Growth scale factor (GSF) Difference in GSF (significant* if difference >8, p < 0.1) 78 7 73 98 45 99 99 47 88 105 63 98 94 34 82 88 21 78 2500+ words (receptive) English Hindi Mandarin Indonesian Total words % Change in total words 984 15 – – 999 2500+ words (expressive) English Hindi Mandarin Indonesian Total words % Change in total words 813 13 – – 826 26* 1177 32 – – 1209 881 – 17 5 903 +21 S2 S3 Pre Post Pre Post Pre Post 53 92 39 69 53 100 50 83 53 61 3 22 53 66 5 31 19 59 2 32 53 73 11 42 +20.3 +40.9 +31.3 independent audit of 10% of therapy videos indicated that the therapy protocol had been adhered to 100% of the time. 3.4.2. Transcription reliability 10% of the GFTA-2 samples were randomly reviewed by an independent Speech Language Pathologist for inter-rater and intra-rater reliability in phonetic transcription results. Intra-rater reliability for transcription data was 92.5% and inter-rater reliability was 83.3%. 4. Discussion 4.1. Summary of findings This study examined whether speech perception intervention was effective for preschool children with hearing loss, and whether intensive intervention for speech perception alone had an impact on vocabulary and speech development. In order to do this, the Speech Perception Education and Assessment Kit (SPEAK) was developed and trialled. A single case, multiple baseline study was implemented with three participants aged 2:6–3:0 years at the start of the programme. The SPEAK-Intervention programme was deemed to be a promising tool for developing speech perception skills in the three young children with all participants progressing through the programme. Results from the ESP as well as the SPEAK-Baseline and Probes showed that speech perception intervention led to positive change for all 3 preschool children with hearing loss. Participants generalised their auditory development to untrained words at the same level as those trained. Moreover, the participants maintained their speech perception skills after intervention was 2322 – 33 8 2363 374 – 17 5 396 +18.6 S1 4 30 – 0 – 30 +161 948 32 – – 980 Table 4 GFTA-2 and articulation percentage consonants correct (PCC) results. Total words attempted GFTA-2 standard score GFTA-2 percentile rank Percentage consonants correct (PCC) Change in PCC (%) 10* +210 1207 – 33 8 1248 +215 89 – 4 – 93 20 – 0 – 20 32 – 4 – 36 +80 withdrawn. The SPEAK-Intervention program is one of the first speech perception intervention programmes demonstrated to be promising for use with preschool children with hearing loss. In addition to progress in speech perception skills, SPEAKIntervention appeared to facilitate vocabulary and speech development. The development of vocabulary and speech as a secondary impact of speech perception intervention in children with hearing loss is an important novel finding. 4.2. Speech perception Speech perception skills of all participants progressed during the SPEAK-Intervention program as demonstrated when assessed using the SPEAK-Baseline in both the total score and skill level. These results indicate that speech perception programmes can be designed and used efficaciously with preliterate pre-schoolers. The success of the programme was not only evident throughout the intervention tasks of the programme, but also in the participants’ ability to generalise their newfound skills to learning unfamiliar or untrained words. This has positive implications for clinicians, as they may not need to teach all words in direct therapy, when using an intensive auditory training programme such as SPEAKIntervention. The results also highlight that the improved speech perception skills are maintained once intervention is withdrawn. The participants’ ability to generalise and maintain speech perception skills from words used in the intervention programme to unfamiliar words is a clinically valuable finding. Implementing programmes that ensure progress is maintained over time provides clinician and families with confidence that their time in intervention is utilised efficiently and effectively. 4.3. Vocabulary In addition to progress in speech perception skills, participants made progress in their vocabulary development following SPEAKIntervention. This change was measured using the PPVT-4 GSV scores which examines if change in vocabulary scores is beyond what would be expected as part of a child’s typical vocabulary development. According to the PPVT-4, a change in GSV score of greater than 8 points shows a significant improvement in vocabulary development beyond the expected from maturity. J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 1723 Table 5 CASALA phonological processes analyses. Age eliminated Initial consonant deletion Alveolar backing Other backing Cluster deletion Devoicing Final consonant deletion Stopping fricatives Cluster reduction Total processes > 20% X X X X 3:0 3:3 3:6 4:0 S1 (%) S2 (%) S3 (%) Pre Post Pre Post Pre Post – 6 2 2 8 19 13 8 – 2 2 – 6 2 4 9 21 25 11 23 23 42 23 25 – 28 25 11 36 42 34 21 5 5 16 11 5 11 26 32 6 9 13 4 9 25 25 26 0 0 7 6 2 3 X indicates infrequently used phonological processes in typically developing children. Two of the 3 participants made statistically significant progress on the PPVT-4 and all increased their receptive and expressive vocabulary, as reported by their parents. The two participants who made significant progress as recorded on the PPVT-4 had reached a higher level on the SPEAK-Intervention programme than the remaining participant, supporting the literature that found correlation between good speech perception skills and vocabulary development [49]. The post-intervention progress of the participants’ expressive vocabulary development as recorded on 2500+ Words was between 18.6% and 215%, although it could not be determined whether the reported increases were statistically significant. There was discrepancy in the degree of vocabulary development recorded on the PPVT-4 with the parent-reported 2500+ words. S1 made the greatest percentile change on the PPVT-4 from 7th to 45th while her parents recorded the lowest number of words being added to her receptive vocabulary at 21%. S2 showed modest development on the PPVT-4 from 47th to 63rd percentile while her parents reported extensive 161% change in her receptive vocabulary. S3 had negative growth on the PPVT-4 from 34th to the 21st percentile on the PPVT, but her parents reported the most change (210%) in her receptive vocabulary. Correlation between parent reports and standardised tests on receptive vocabulary development was examined by Thal [67]. They noted that parent reports for receptive language did not correlate with standardised tests, suggesting that parent reports on a child’s receptive language development need to be supplemented with the collection of objective data. Differences in progress as recorded using standardised tools (such as the PPVT-4) and informal tools are not uncommon. Standardised tests remove context for the child and therefore identifying knowledge that the child has consolidated. In addition, standardised tests often contain culture specific language, which can affect response to stimuli and standardisation of results. For example, some words used in the earliest sections of the PPVT-4 such as ‘cookie’, ‘muffin’, ‘mail’, ‘lamp’, and ‘squirrel’ are likely to be more familiar to children in the norming population. Local children are likely to use the words ‘biscuit’, ‘cake’, ‘letters’, ‘light’, in place of the first 4 words while squirrels are uncommonly seen in Singapore. Informal tools such parent report provide valuable information regarding the child’s skills and knowledge in context and therefore may provide a more accurate representation of the children’s vocabulary development in everyday situations, in particular expressive vocabulary. Parent-reported vocabulary has been found to have strong correlation with standardised tests where studies have been done on families local to the places where the standardised tests were constructed [68,69]. In the absence of locally standardised vocabulary tests, emphasis should therefore be placed on parent-reported measures of vocabulary. It appears that with improved speech perception skills, children with hearing loss can learn words without actively being taught them. This finding has implications for the efficient development of vocabulary in preschool children born with hearing loss who, according to Walker and McGregor [54], continue to experience delay in word learning skills in comparison to their normal hearing peers in spite of early auditory intervention. 4.4. Speech development The speech of all children who were recruited in the study also improved over the short period of SPEAK-Intervention. Progress was made in speech development on the GFTA-2 for 2 of the 3 participants, and on PCC scores for all participants. However, it is unknown as to whether this progress was a result of typical speech development or statistically significant growth. Phonological process development demonstrated by the participants was, for the most part, within the normal range for their age. One participant did not demonstrate progress on the GFTA-3, even though she demonstrated the most improvement on the PCC scores in addition to a reduction in the use of atypical phonological processes. The fact that speech developed over the period of the intervention when speech skills were not actively being taught, is positive in itself. This outcome supports the literature demonstrating that speech perception skills facilitate speech skills in children with hearing loss, and that without good speech perception a child’s speech will be affected. Although there are comparative studies examining the relationship between speech perception skills and speech development [70,71], this is one of the first studies to demonstrate on the impact of auditory intervention using a defined auditory hierarchy, on speech development. This prospective intervention study using speech perception tests scored independent of children’s speech output is therefore an important contribution to the understanding of the impact of speech perception development on speech development. 4.5. Clinical implications The SPEAK-Intervention programme was shown to be effective in developing the speech perception skills of three young preschoolers with hearing loss. Using a predefined set of words in the SPEAK-Intervention programme, the participants not only progressed through the auditory levels, but also generalised these skills to novel words. The ability to generalise speech perception skills learnt to processing unfamiliar words is crucial for children to process and learn from the language in their environment. Therefore SPEAK-Intervention programme may be a useful clinical 1724 J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 tool that can be used to focus a child’s auditory training in an intensive manner for short periods of time during early intervention. It appears that additional benefits of the SPEAK-Intervention programme may include the concomitant development of vocabulary and speech skills, despite these skills not being directly trained. The development of auditory skills to establish speech and language development is the cornerstone of AVT. This is one of the first studies providing preliminary support to the AVT method of developing speech and language skills through listening and an auditory hierarchy. By teaching the children to listen using this method, language can be overheard and learnt without direct teaching of every aspect of speech and language over time [72]. However, it appears that the children need to be at a particular level of the auditory hierarchy for this to occur. Further research is warranted to explore this relationship further. Both S1 and S2 reached criterion at level 10 of the intervention programme and S3 reached level 7 by the end of the study. While all improved their speech skills following SPEAK-Intervention, only the two participants with the higher level of speech perception development had clinically significant improvements on the PPVT4. Therefore, it may be hypothesised that a particular speech perception skill-level threshold is necessary for children to effectively listen to and learn words from their daily activities; and that a lower level of speech perception skill is enough for speech skills to progress without direct speech training. This is one of the first Intervention studies that aims to investigate the efficacy of a specific component of auditory-verbal therapy, and the single-case multiple baseline design has been found to be important for the purposes of further study on the efficacy of components of and techniques used in AVT. Singlesubject study designs are prevalent in the communication sciences literature [73]. Its flexibility enables the researcher to tailor an intervention in accordance to the individualised needs of the participants involved in the study, yielding results that have ready application to clinical practice in addition to providing empirical evidence in support for therapeutic interventions [73]. Given the lack of evidence in documenting the therapeutic process of auditory-verbal practice [30], it is recommended that singlesubject studies be adopted within clinical practice for more widespread efficacy studies of therapeutic strategies used in auditory-verbal practice. perception skills in young children with hearing loss should not completely replace functional speech and language intervention. Speech and language skills are learnt through play and social interaction, and should not be forgotten. It is hoped through further studies using the SPEAK-Intervention programme, the authors can find the ideal mix between functional therapy and intensive speech perception intervention to produce the most effective and efficient early intervention programme for optimal speech and oral language development for children with hearing loss. 5. Conclusions This is one of the first intervention studies to respond to the call of Moeller et al. [7] to determine whether focussed listening practice can impact on speech development. The results of this Phase 1 study shows promise regarding training speech perception and the impact on speech development. However, further research with a larger number of children and in particular the inclusion of control children is necessary. More specifically, the study examined the efficacy a newly developed speech perception training programme (SPEAK-Intervention), and the impact of training speech perception on the vocabulary and speech development of young preschool aged children with hearing loss. Our findings provide preliminary support for the effective provision of auditory intervention to preschool children with permanent hearing loss. Results showed that speech perception training is possible with young children and that this training has some additional benefits for vocabulary and speech development. This in-turn makes this intervention study one of the first to demonstrate the premise of AVT; supporting the development speech and language through listening. However, findings should be viewed with caution as the study needs to be replicated on a larger cohort. In addition, some research into optimal practice schedules using the SPEAK-Intervention programme and where it fits in the typical early intervention process are required. Acknowledgements The authors wish to thank colleagues at the ENT Clinic, National University Hospital, as well as the families who participated in this study. 4.6. Limitations References The findings from the study are exciting however there are several limitations to this study. The participant numbers was small and therefore outcomes cannot be directly generalised to the population of young hearing impaired children. Further controlled studies with a larger cohort are required. The authors were unable to run statistics on the data due to issues with autocorrelation. This will need to be considered in the development of future methodologies, as determining whether progress in post-intervention outcomes is significant or otherwise is required. The lack of normative vocabulary and speech development data on Singaporean pre-school-aged children affected the extent to which data collected may be compared to children with normal hearing in Singapore. It was also difficult to accurately ascertain the level of vocabulary or speech perception development for the children recruited in this study. Future Phase-2 studies using SPEAK-Intervention should include a control group to allow the influence of developmental growth to be measured more accurately. Finally, the results should be viewed with caution. While the SPEAK-Intervention programme has promise, drilling speech [1] Australian Hearing, Demographic Details and Aetiology of Persons Under the Age of 17 Years With a Hearing Impairment Who Have Been Fitted With a Hearing Aid, National Acoustic Laboratories, Chatswood, 2005. [2] H.M. Fortnum, A.Q. Summerfield, D.H. Marshall, A.C. Davis, J.M. Bamford, Prevalence of permanent childhood hearing impairment in the United Kingdom and implications for universal neonatal hearing screening: questionnaire-based ascertainment study, Br. Med. J. 323 (2001) 536–540. [3] A.L. Mehl, V. Thomson, Newborn hearing screening: the great omission, Pediatrics 101 (1) (1998) e4. [4] National Center for Hearing Assessment and Management, Prevalence of Congenital Hearing Loss, NCHAM, 2010 Available from: http://www.infanthearing.org/summary/prevalence.html (accessed 07.03.13). [5] M.P. Moeller, E. McCleary, C. Putman, A. Tyler-Krings, B. Hoover, P. Stelmachowicz, Longitudinal development of phonology and morphology in children with lateidentified mild–moderate sensorineural hearing loss, Ear Hear. 31 (5) (2010) 625–635. [6] M.P. Moeller, J.B. Tomblin, C. Yoshinaga-Itano, C.C. McDonald, S. Jerger, Current state of knowledge: language and literacy of children with hearing impairment, Ear Hear. 28 (2007) 740–753. [7] M.P. Moeller, B. Hoover, C. Putman, K. Arbataitis, G. Bohnenkamp, B. Peterson, et al., Vocalizations of infants with hearing loss compared with infants with normal hearing: Part I. Phonetic development, Ear Hear. 28 (5) (2007) 605–627. [8] J. Lukomski, Deaf college students’ perceptions of their social–emotional adjustments, J. Deaf Stud. Deaf Educ. 12 (4) (2007) 486–494. [9] M.P. Moeller, Current state of knowledge: psychosocial development in children with hearing impairment, Ear Hear. 28 (6) (2007) 729–739. J. Lew et al. / International Journal of Pediatric Otorhinolaryngology 78 (2014) 1716–1725 [10] P. Theo van Eldik, J.W.V. Treffers, F.C. Verhulst, Mental health problems of deaf Dutch children as indicated by parents’ responses to the child behavior checklist, Am. Ann. Deaf 148 (5) (2004) 390–395. [11] L. Robertson, Literacy Learning for Children Who Are Deaf of Hard of Hearing, A.G. Bell Association for the Deaf and Hard of Hearing, Washington, 2000. [12] A.E. Geers, H. Hayes, Reading, writing, and phonological processing skills of adolescents with 10 or more years of cochlear implant experience, Ear Hear. 32 (1) (2010) 49S–59S. [13] F. Venail, A. Vieu, F. Artieres, M. Mondain, A. Uziel, Educational and employment achievements in prelingually deaf children who receive cochlear implants, Arch. Otolaryngol. Head Neck Surg. 136 (4) (2010) 366–372. [14] Cooperative Research Centre for Cochlear Implant and Hearing Aid Innovation (CRC HEAR), Listen Hear: The Economic Impact and Cost of Hearing Loss in Australia Access Economics, VicDeaf, Victoria, 2006. [15] D. Jung, N. Bhattacharyya, Association of hearing loss with decreased employment and income among adults in the United States, Ann. Otol. Rhinol. Laryngol. 121 (12) (2012) 771–775. [16] P.E. Mohr, J.J. Feldman, J.L. Dunbar, A. McConkey-Robbins, J.K. Niparko, R.K. Rittenhouse, et al., The societal costs of severe to profound hearing loss in the United States, Int. J. Technol. Assess. Health Care 16 (4) (2000) 1120–1135. [17] E. Cole, C. Flexer, Children with Hearing Loss: Developing Listening and Talking (Birth to Six), Plural Publishing, San Diego, 2008. [18] D. Querleu, X. Renard, F. Versyp, L. Paris-Delrue, G. Crepin, Fetal hearing, Eur. J. Obstet. Gynecol. Reprod. Biol. 29 (1988) 191–212. [19] A. Vouloumanos, J.F. Werker, Listening to language at birth: evidence for a bias for speech in neonates, Dev. Sci. 10 (2) (2007) 159–171. [20] F. Ramus, M.D. Hauser, C. Miller, D. Morris, J. Mehler, Language discrimination by human newborns and by cotton-top tamarin monkeys, Science 288 (5464) (2000) 349–351. [21] R.J. Ruben, R. Schwartz, Necessity versus sufficiency: the role of input in language acquisition, Int. J. Pediatr. Otorhinolaryngol. 47 (2) (1999) 119–226. [22] J. Mehler, P. Jusczyk, G. Lambertz, N. Halsted, J. Bertoncini, C. Amiel-Tison, A precursor of language acquisition in young infants, Cognition 29 (2) (1988) 143–178. [23] D.L. Molfese, V.J. Molfese, Hemisphere and stimulus differences as reflected in the cortical responses of newborn infants to speech stimuli, Dev. Psychol. 15 (6) (1979) 505–511. [24] P.G. Simos, D.L. Molfese, Electrophysiological responses from a temporal order continuum in the newborn infant, Neuropsychologia 35 (1) (1997) 89–98. [25] A. Sharma, A.A. Nash, M. Dorman, Cortical development, plasticity and reorganisation in children with cochlear implants, J. Commun. Disord. 42 (2009) 272–279. [26] Joint Committee on Infant Hearing, Year 2000 position statement: principles and guidelines for early hearing detection and intervention programs, Pediatrics 106 (2000) 798–817. [27] Joint Committee on Infant Hearing, Year 2007 position statement: principles and guidelines for early hearing detection and intervention programs, Pediatrics 120 (4) (2007) 898–921. [28] Centers for Disease Control and Prevention, National Center for Health Statistics, Healthy People 2010: Final Review, 2011. [29] S. Schwartz, Choices in Deafness: A Parents’ Guide to Communication Options, 2007. [30] E.A. Rhoades, Research outcomes of auditory-verbal intervention: is the approach justified? Deaf. Educ. Int. 8 (3) (2006) 125–143. [31] A.G. Bell Academy for Listening and Spoken Language, Principles of AuditoryVerbal Therapy, 2007 Available from: http://listeningandspokenlanguage.org/ AcademyDocument.aspx?id=563 (accessed 18.01.13). [32] W. Estabrooks, Auditory-Verbal Therapy and Practice, Alexander Graham Bell Association for the Deaf and the Hard of Hearing Inc., USA, 2006. [33] D. Dornan, L. Hickson, B. Murdoch, T. Houston, G. Constantinescu, Is auditoryverbal therapy effective for children with hearing loss? Volta Rev. 110 (3) (2010) 361–387. [34] E.A. Rhoades, T.H. Chisolm, Global language progress with an auditory-verbal approach for children who are deaf or hard of hearing, Volta Rev. 102 (1) (2000) 5–25. [35] A. Fulcher, A.A. Purcell, E. Baker, N. Munro, Listen up: children with early identified hearing loss achieve age-appropriate speech/language outcomes by 3 years-of-age, Int. J. Pediatr. Otorhinolaryngol. 76 (12) (2012) 1785–1794. [36] D. Dornan, L. Hickson, B. Murdoch, T. Houston, Outcomes of an auditory-verbal program for children with hearing loss: a comparative study with a matched group of children with normal hearing, Volta Rev. 107 (1) (2006) 37–54. [37] J. Tuohy, J. Brown, C. Mercer-Moseley, L. Walsh, St. Gabriel’s Curriculum, second ed., St Gabriel’s School for Hearing Impaired Children, Sydney, 2005. [38] Cochlear, Listen Learn and Talk, SOS Printing Group, Australia, 2003. [39] Q.-J. Fu, J. Galvin, X. Wang, G. Nogaki, Moderate auditory training can improve speech performance of adult cochlear implant patients, Acoust. Res. Lett. Online 6 (3) (2005) 106–111. [40] G.C. Stecker, G.A. Bowman, E.W. Yund, T.J. Herron, C.M. Roup, D.L. Woods, Perceptual training improves syllable identification in new and experienced hearing aid users, J. Rehabil. Res. Dev. 43 (4) (2006) 537–552. [41] R. Sweetow, J.H. Sabes, The need for and development of an adaptive Listening and Communication Enhancement (LACE) program, J. Am. Acad. Audiol. 17 (2006) 538–558. 1725 [42] J.-L. Wu, H.-M. Yang, Y.-H. Lin, Q.-J. Fu, Effects of computer-assisted speech training on Mandarin-speaking hearing-impaired children, Audiol. Neurotol. 12 (2007) 307–312. [43] D. Swingley, The roots of the early vocabulary in infants’ learning from speech, Curr. Dir. Psychol. Sci. 17 (5) (2008) 308–312. [44] J.F. Werker, C.T. Fennell, Infant speech perception and language acquisition: methodological underpinnings, in: J. Colombo, P. McCardle (Eds.), Infant Pathways to Language: Methods, Models, and Research Directions, Taylor & Francis Group, New York, 2009, pp. 85–98. [45] J.F. Werker, H.H. Yeung, Infant speech perception bootstraps word learning, Trends Cogn. Sci. 9 (11) (2005) 519–527. [46] P. Kuhl, Linking infant speech perception to language acquisition, in: P. McCardle, J. Colombo, L. Freund (Eds.), Infant Pathways to Language: Methods, Models and Research Directions, Taylor & Francis Group, New York, 2009, pp. 213–244. [47] D.L. Molfese, Predicting dyslexia at 8 years of age using neonatal brain responses, Brain Lang. 72 (2000) 238–245. [48] D.L. Molfese, V.J. Molfese, Discrimination of language skills at five years of age using event related potential recorded at birth, Dev. Neuropsychol. 24 (1997) 541–558. [49] J.L. Desjardin, S.E. Ambrose, A.S. Martinez, L.S. Eisenberg, Relationships between speech perception abilities and spoken language skills in young children with hearing loss, Int. J. Audiol. 48 (2009) 248–259. [50] L.E. Paatsch, P.J. Blamey, J.Z. Sarant, C.P. Bow, The effects of speech production and vocabulary training on different components of spoken language performance, J. Deaf Stud. Deaf Educ. 11 (1) (2006) 39–55. [51] D.L. Mills, C. Prat, R. Zangl, C.L. Stager, H.J. Neville, J.F. Werker, Language experience and the organisation of brain activity to phonetically similar words: ERP evidence from 14- to 20-month-old, J. Cogn. Neurosci. 16 (2004) 1452–1464. [52] C.L. Stager, J.F. Werker, Infants listen for more phonetic detail in speech perception than in word-learning tasks, Nature 388 (1997) 381–382. [53] J.F. Werker, C.T. Fennell, K.M. Corcoran, C.L. Stager, Infants’ ability to learn phonemically similar words: effects of age and vocabulary size, Infancy 3 (1) (2002) 1–30. [54] E.A. Walker, K.K. McGregor, Word learning processes in children with cochlear implants, J. Speech Lang. Hear. Res. 56 (2) (2013) 375–387. [55] A.R. Lederberg, P.E. Spencer, Word-learning abilities in deaf and hard-of-hearing preschoolers: effect of lexicon size and language modality, J. Deaf Stud. Deaf Educ. 14 (1) (2009) 44–62. [56] B. Dodd, B. McIntosh, Two-year-old phonology: impact of input, motor and cognitive abilities on development, J. Child Lang. 37 (5) (2010) 1027–1046. [57] L.S. Eisenberg, Current state of knowledge: speech recognition and production in children with hearing impairment, Ear Hear. 28 (6) (2007) 766–772. [58] A. Boothroyd, Speech perception and sensorineural hearing loss, in: M. Ross, T.G. Giolas (Eds.), Auditory Management of Hearing-Impaired Children, University Park Press, Baltimore, 1978, pp. 117–144. [59] J.S. Moog, A.E. Geers, Early Speech Perception (ESP) Test, Central Institute for the Deaf, St. Louis, 1990. [60] L.M. Dunn, D.M. Dunn, Peabody Picture Vocabulary Test (PPVT-4), fourth ed., Pearson, Texas, 2007. [61] S. Armstrong, 2500+ Words List, St. Gabriele’s Early Intervention Centre, Brisbane, 1992. [62] R. Goldman, M. Fristoe, Goldman Fristoe Test of Articulation (GFTA), second ed., American Guidance Service, USA, 2000. [63] L.D. Shriberg, D. Austin, The Percentage of Consonants Correct (PCC) metric: extensions and reliability data, J. Speech Lang. Hear. Res. 40 (4) (1997) 708–722. [64] H.E.A. Rworks, Computer Aided Speech and Language Assessment (CASALA), HEARworks, Australia, 2007. [65] L.V. McReynolds, M. Elbert, Criteria for phonological process analysis, J. Speech Lang. Hear. Res. 46 (2) (1981) 197–204. [66] J.E. Roberts, M. Burchinal, M.M. Footo, Phonological processes decline from 2.5 to 8 years, J. Commun. Disord. 23 (3) (1990) 205–217. [67] D.J. Thal, L. O’Hanlon, M. Clemmons, L. Fralin, Validity of a parent report measure of vocabulary and syntax for preschool children with language impairment, J. Speech Lang. Hear. Res. 42 (2) (1999) 482–496. [68] A.K. Prezbindowski, A.R. Lederberg, Vocabulary assessment of deaf and heard-ofhearing children from infancy through the preschool years, J. Deaf Stud. Deaf Educ. 8 (4) (2003) 383–400. [69] D. Thal, J.L. Desjardin, L.S. Eisenberg, Validity of the MacArthur–Bates Communicative Development Inventories for measuring language abilities in children with cochlear implants, Am. J. Speech Lang. Pathol. 16 (2007) 54–64. [70] P.J. Blamey, J.Z. Sarant, L.E. Paatsch, J.G. Barry, C.P. Bow, R.J. Wales, et al., Relationships among speech perception, production, language, hearing loss and age in children with impaired hearing, J. Speech Lang. Hear. Res. 44 (2) (2001) 264–285. [71] L.E. Paatsch, P.J. Blamey, J.Z. Sarant, L.F.A. Martin, C.P. Bow, Separating contributions of hearing, lexical knowledge, and speech production to speech-perception scores in children with hearing impairments, J. Speech Lang. Hear. Res. 47 (4) (2004) 738–750. [72] N. Akhtar, J. Jipson, M.A. Callanan, Learning words through overhearing, Child Dev. 72 (2) (2001) 416–430. [73] R.L. Tate, S. Mcdonald, M. Perdices, L. Togher, R. Schultz, S. Savage, Rating the methodological quality of single-subject designs and n-of-1 trials: introducing the Single-Case Experimental Design (SCED) scale, Neuropsychol. Rehabil. 18 (4) (2008) 385–401.