Abstract
This case study described the oral expressive outcomes of five children with hearing loss who experienced prolonged auditory deprivation prior to participating in an auditory intervention. Expressive outcomes were measured by the number of spontaneous words and imitations. Visual analyses revealed that two of the five participants increased their oral expressive skills during the intervention. The results suggest that children with hearing loss who are late identified and aided can still develop some basic oral expressive skills, but they would probably benefit from interventions that directly target the use of words expressively.
Children with hearing loss who experience a period of auditory deprivation longer than approximately 3.5 years of age do not seem to develop auditory and oral expressive outcomes like hearing peers (e.g., Sharma et al., 2002). During these first 3 years of life, the brain shows maximal plasticity and early auditory exposure is crucial for the development of specialized auditory cortical areas in the brain. Previous studies have shown than if the brain is not stimulated with auditory stimuli, the auditory cortical areas will be recruited to process visual or somatosensory stimuli (e.g., Kral et al., 2016; Sharma et al., 2005). This process is known as brain reorganization and may be responsible for the limited auditory and oral expressive outcomes of some late-identified and aided children with hearing loss. Although there is plasticity throughout life to some extent (Morishita & Hensch, 2008), studies have suggested that brain plasticity is greatly reduced around 7 years of age (Dorman et al., 2007; Gilley et al., 2008; Sharma et al., 2015). Late identification and amplification are common in developing countries where access to hearing technology and speech and hearing services is not always possible. One such as case is Nicaragua.
A Closer Look at Deafness in Nicaragua
The rate of congenital severe to profound sensorineural hearing loss in Nicaragua is much higher than in developed countries, with an estimated school-age prevalence of 1.6 to 7.2 per 1,000 compared with 1 to 3 per 1,000 in developed countries (e.g., Olusanya & Newton, 2007; Pascolini & Smith, 2009). A survey of national disabilities in Nicaragua indicated that 12% of those with disabilities had a hearing loss (Saunders et al., 2015). Because there is no universal newborn hearing screening, children with hearing loss are identified when they do not learn to talk (Saunders et al., 2015); many children go untreated and undiagnosed until school entry when they live in rural areas and with limited access to medical care.
Most children with hearing loss in Nicaragua (66%) receive no services at school (Saunders et al., 2015). Children who are deaf or hard of hearing (DHH) arrive to school between the ages of 5 and 7 years with nothing more than a few home signs. A minority of children with hearing loss receives amplification and that is typically inconsistent due to frequent breakdowns, difficulties in maintaining a supply of hearing aid batteries (Polich, 2001), and accessing services near their homes. The amplification is not followed necessarily by appropriate parent training or educational practices. There are few professionals in the country trained in audiology or speech-language pathology (Polich, 2001) because the universities do not offer training in these areas yet.
In 1979, special education was included as part of the Nicaragua’s educational reform, and, prior to that, any special education schools were private (Polich, 2001). Nicaraguan Sign Language (NSL) emerged naturally in the schools for the deaf where children lived, being one of the most recent languages created (Senghas & Coppola, 2001). By 1980, schools became part of the public education system, and integration with regular education was established. However, children who are DHH are difficult to integrate into the regular education system. In 1992, the National Nicaragua Association for the DHH started to work with the Ministry of Education to provide sign language instruction. Currently, teachers who are fluent signers can be assigned to teach this population and there are more self-contained and special schools for the deaf in the country using NSL than in previous decades. Hearing technology and speech and hearing services are very limited even when children have residual hearing and they could actually benefit from them.
Expressive Outcomes After Auditory Deprivation
Research with children who have not acquired language due to auditory and/or environmental deprivation indicates that the deprivation can cause significant language delays (e.g., Fromkin et al., 1974; Friedmann & Rusou, 2015). In addition, previous studies indicate that the acquisition of grammar is difficult, especially in the area of syntax (Friedmann & Rusou, 2015; Friedmann & Szterman, 2011). For example, Friedmann and Szterman (2005) studied the comprehension and production of sentences derived by syntactic movement in 20 Hebrew-speaking children with moderate to profound hearing loss between 7 and 10 years of age. Children were matched to a hearing control group based on the language level. They found that children with hearing loss failed to understand complex sentences, such as object relatives and object-verb-subject topicalization sentences. In the production tasks, children with hearing loss either avoided producing a sentence with syntactic movement or produced ungrammatical sentences. The authors concluded that both comprehension and production of children with hearing loss were significantly different from that of the hearing control group and that only children who received hearing aids before the age of 8 months performed well in the comprehension tasks.
Studies examining early identification of hearing loss also indicate that those who are identified by 6 months of age and receive early intervention develop higher language skills than those identified and intervened later (e.g., de Diego-Lázaro et al., 2019; Yoshinaga-Itano et al., 1998, 2017). For example, Yoshinaga-Itano et al. (2017) found that children who met the Early Hearing Detection and Intervention (EHDI) policy guidelines had better vocabulary outcomes than those who did not meet the EHDI guidelines in a sample of English-speaking children in the U.S. EHDI guidelines recommending hearing loss screening by 1 month, identification by 3 months, and amplification and enrollment in intervention by 6 months of age (American Academy of Pediatrics, 2007). Similarly, de Diego-Lázaro and colleagues found that the age of enrollment in early intervention was a significant predictor of vocabulary outcomes in a group of Latino children in the United States, whose home language was Spanish. Although these studies highlight the importance of early hearing loss identification and intervention for language outcomes, it is unknown what expressive outcomes children in developing countries, where early identification and intervention are not part of the standard protocol, can develop. The purpose of this case study was to describe the oral expressive outcomes of children with hearing loss who had a long period of auditory deprivation while they received an auditory intervention.
Method
Participants
Children participating in this study came from a larger project investigating the effect of an auditory intervention on the auditory skills of children with hearing loss who received hearing aids after 7 years of age (de Diego-Lázaro & Restrepo, in press). Children were originally from remote rural areas in Nicaragua and were living in a group home at the time of the study. Twenty children from 6 to 18 years of age lived in the group home with five caregivers. Caregivers and children used NSL to communicate with each other. Children attended a school nearby where they received education in NSL. Before enrollment in the group home, children had 1 to 2 years of previous education in regular schools (preschool or first grade), no previous exposure to NSL, and no previous use of hearing aids nor speech and language services. Their ability to communicate with others was limited to some basic homemade signs and lip reading of basic words.
Specific eligibility criteria for the study included children (a) with bilateral sensorineural moderate to profound hearing loss, (b) who received hearing aids after the sensitive period for auditory skills development (after 7 years of age), (c) who used hearing aids all waking hours, (d) who showed an aided pure tone average of 60 dB or better, and (e) without additional disabilities per caregiver and teacher report. Five children, three males and two females, between 9 and 12 years of age participated in this study. Table 1 shows the participant’s characteristics at the beginning of the intervention. Participant 1 experienced gentamicin overdose right after birth. The cause of hearing loss was unknown for the other four participants.
Participant Characteristics.
Note. Age and length of hearing aid use is expressed in years. PTA = pure tone average; dB HL = decibels hearing level; IT MAIS: Infant-Toddler Meaningful Auditory Integration Scale (maximum of 40 points); pragmatic skills maximum of 11 points.
Pure-tone average for 500; 1,000; and 2,000 Hz in better ear.
When children enrolled in the group home, they received a complete hearing evaluation comprising an otoscopy, tympanometry, pure tone audiometry, and speech detection test. Children were fitted with bilateral hearing aids, Resound Digital Mach 70, 80 or 90, based on the hearing evaluation. All the participants had residual hearing and benefited from the use of hearing aids as shown by the aided speech detection thresholds (see Table 1). Children entered the group home at different points in time, and consequently, their exposure to NSL, length of hearing aid use, and pragmatic skills varied. All the children had less than 50 spoken words.
Given the unique characteristics of the children included in the study and the impossibility of using standardized assessments to characterize the sample, we asked teachers and group home caregivers to complete two surveys to assess the auditory and language skills of the participants before the intervention. Teachers and caregivers completed the Infant-Toddler Meaningful Auditory Integration Scale (IT-MAIS; Zimmerman-Phillips et al., 1997) to characterize the participant’s auditory skills and a language checklist developed for this study. The IT-MAIS is designed to assess the child’s spontaneous responses to sound in his or her everyday environment. The assessment is based on information provided by the child’s caregivers in response to 10 questions regarding vocalization behavior, alerting to sounds, and deriving meaning from sound. The language checklist included questions about the children’s number of signs, spoken words, and pragmatic skills (using either sign or spoken language). Pragmatic skills included questions such as asking for information or objects, commenting, greetings, or negation (Halliday, 1975).
Intervention Procedures
Participants received 27 individual intervention sessions distributed in three sessions per week for 9 weeks. The goal of the intervention was to increase children’s auditory skills in Spanish. Although participants were given turns during the intervention, they were not required to repeat the words or practice them expressively. The sessions were 35 minutes long and followed the same structure: listening check (checking hearing aids and Ling test), probe (every three sessions), review of previously taught words, play-based activities, and a narrative activity. Play-based activities included puzzles, coloring and writing activities, and preparing meals. The narrative activity consisted of drawing a sequence while creating a short story with the target words. The local provider helped create the story and ensured that all the stories had a linear structure of exposition, conflict, and resolution. She asked leading questions to elicit the story from the children (e.g., “what happened to the child?” “How is the child feeling?” “What can he/she do?” etc.). Each session focused on words from one theme (food, body parts, feelings, and clothing). A full description of the intervention and the targeted words can be found in de Diego-Lázaro and Restrepo (in press).
Teletraining of a Local Provider
A local provider was teletrained to provide the intervention. She was a 28-year-old medical doctor from Nicaragua 1 . She was trained as an audiology technician by a U.S. medical mission and she practiced part-time in a local hospital. She was fluent in NSL. The trainer (the first author on this article) was a speech and language pathologist and teacher of the deaf with 6 years of experience in auditory-verbal therapy (AVT). The trainer was in the United States during the training. The online training contained three phases (see Figure 1) and lasted 6 months.
Tele-training phases.
Phase 1, concept learning, occurred before delivering the intervention and consisted of six meetings using Skype software. The meetings lasted for 90 minutes and were distributed across 2 months. The local provider received readings and videos on each session’s topic ahead of time that were discussed during the online meetings. Meetings 1 through 3 focused on auditory and language development, communication options, and informal and dynamic assessments. Meetings 4 and 5 focused on auditory and language support strategies derived from AVT (Rhoades et al., 2016). The local provider was introduced to all AVT principles, but she was coached to use strategies that prioritized auditory over visual information and that increased the number of times children heard each word. These strategies were auditory first (e.g., naming an object a few times before showing it), auditory sandwich (presenting information auditorily, using a visual cue such as lip reading or a picture if necessary, and auditorily again), interpretation (translating children’s signs into spoken language), parallel talk (describing what the child is doing or seeing), and self-talk (describing what the provider is doing or seeing). During the last meeting, the local provider learned about the goal, design, and logistics of the intervention, including how to use Vydeo software. The provider did not receive training on audiology, given her previous training as an audiology technician. All the training materials and meetings were given in Spanish.
Phase 2, real-time coaching, lasted for 9 weeks. The local provider received coaching in real time while providing the intervention to Participants 2, 4, and 5. We used Vydeo software to provide the real-time coaching in which the trainer camera was off but the microphone was still on. That way, we could watch and coach the provider being minimally distracting to the children. For example, we could remind the local provider to describe what the child was doing during the activity (i.e., parallel talk), without being watched by the children. In addition, the provider was required to record one session per week for each participant. Sessions were recorded using the laptop camera. Once a week, the provider and the trainer met online during 30 minutes to go over the goals, strategies, and activities appropriate for each child.
During Phase 3, delayed feedback, we removed the real-time coaching and the provider only received delayed feedback once per week based on the recorded videos for Participants 1 and 3. The sessions were recorded whenever electricity supply was available. The videos were used to assess intervention fidelity and outcome measures for the participant’s expressive skills. All the training materials and videos were shared through Arizona State University’s protected Dropbox folder. Arizona Dropbox system and Vydeo software are Health Insurance Portability and Accountability Act (HIPAA) compliant.
Measures
Expressive outcomes
For every participant, we transcribed the narrative activity for every available video during the intervention phase. We transcribed using in T-units (i.e., shortest allowable grammatical units). We chose the narrative activity because it was typically the least structured activity in the session, and thus, children used more words spontaneously. We coded children’s number of oral word imitations, spontaneous words, and mean length utterance in words using Systematic Analysis of Language Transcripts (SALT) software (Miller & Chapman, 2008). Utterances could be oral only (Spanish) or oral and NSL simultaneously. For word imitation, we distinguished between direct and spontaneous imitations. In direct imitations, the local provider gave a turn to the child (e.g., she asked a question, modeled the answer, and the child imitated), she looked at the child expectantly, or verbally indicated that a response was necessary (e.g., say “hi”). For spontaneous imitation, the child imitated the local provider without being asked for it. For spontaneous words, we considered intelligible words produced by the child that were not used or prompted by the local provider in five preceding turns.
Interobserver agreement
The first author transcribed and coded all the narrative activities in the available intervention videos for each of the children. A second trained observer (e.g., an undergraduate research assistant) analyzed 35% of the videos randomly selected and recoded the transcriptions independently from the original scoring. The totals of each code per session (direct imitations, spontaneous imitations, and spontaneous words) were compared across observers for agreement. The total of agreements was divided by the total number of comparison points and multiplied by 100 in each measure. Interobserver reliability for direct imitations was between 89% and 96%, for spontaneous imitations was 90% and 95%, and for spontaneous words was 88% and 93%.
Results
Of the 27 planned intervention sessions, Participant 2 received 24 sessions, Participant 3 received 22, and Participants 1, 4, and 5 received 26. For Participants 2, 4, and 5, the local provider was able to record 10 to 11 intervention sessions. For Participants 1 and 3, she was able to record eight and seven sessions, respectively, due to shutdowns in the electricity supply needed to use the laptop camera. Participants’ percentage of utterances with word imitations (direct and spontaneous) and spontaneous words during the narrative activities were plotted in Figure 2 for visual analyses. The rest of the utterances up to 100% in the graphs were unintelligible or nonverbal utterances in NSL. Among the participants who communicated expressively during the intervention (Participants 2, 4, and 5), we observed high variability in their production of imitations and spontaneous words from one session to the other. This was because the theme and the activities children did in every session were different, and thus, their motivation to participate and communicate also varied.
Participants 1 to 5 percent of utterances with imitations and spontaneous words during the intervention.
Participant 1 showed a flat pattern of spontaneous imitations between 0 and 10 throughout the intervention. Participant 1 did not produce any word spontaneously during the intervention and his mean length of utterance (MLU) was one word. Overall, only 5% of the utterances he produced contained word imitations. Ninety-five percent of the utterances were signs that he copied from the interventionist because he was learning NSL when he received the intervention.
Participant 2 showed increases in his use of spontaneous imitations and words. His number of spontaneous and direct imitations varied throughout the intervention phase based on his motivation with the activities. Average of direct imitations was 15 and average of spontaneous imitations was 19. Participant 2 showed an MLU of 1.14 words. On average, Participant 2 communicated verbally 43% of the time, with or without the use of sign, whereas 57% of the time he communicated using sign language only.
Participant 3 did not communicate verbally during the intervention, showing a flat pattern of spontaneous and direct imitations, and spontaneous words. In the last session, Participant 3 imitated the interventionist 17% of the time after request and 8% spontaneously. Participant 3 showed an MLU of one word. Overall, only 5% of the utterances that she produced contained word imitations. Ninety-five percent of the utterances were signs in NSL.
Participant 4 showed a steady production of direct and spontaneous imitations throughout the intervention (around 20% of the time), with the exception of Session 16 where she only produced spontaneous imitations 5% of the time and direct imitations 9% of the time. Her production of spontaneous words was also steady throughout the intervention, below 9% of the time. Participant 4 showed an MLU of 1.05 words. On average, she communicated verbally 35% of the time, with or without the use of sign, whereas 65% of the time she communicated using sign language only.
Participant 5 showed the greatest increase in his production of spontaneous words, from 4% of the time in Session 1 to 62% of the time in Session 26. His production of spontaneous imitations also increased during the intervention from 29% of the time in Session 1 to 62% of the time in Session 26. Participant 5 production of direct imitations was steady during the intervention and below 7% of the time. The provider did not need to request for direct imitations as frequent as for the other participants in the study because Participant 5 was producing and imitating words spontaneously. He showed an MLU of 1.59 words. On average, he communicated verbally 98% of the time, with or without the use of sign, whereas 2% of the time, he communicated using sign language only.
Discussion
The purpose of this study was to describe the oral expressive outcomes of children with hearing loss who had a long period of auditory deprivation while they received an auditory intervention. Participants 1, 3, and 4 exhibited a flat pattern in their use of imitations and spontaneous words during the intervention. Participant 2 showed increases in spontaneous and direct imitations, and Participant 5 showed increases in spontaneous words and imitations. Therefore, being in a rich language environment (i.e., the auditory intervention) motivated some children to use their expressive skills, but not all of them. The results suggest that children with hearing loss who are late identified and aided may require direct speech intervention to develop oral expressive skills.
There are few characteristics that may explain why Participants 2 and 5 showed higher oral expressive outcomes than the rest of the participants; they were 2 to 3 years older, had been longer at the group home, had better-aided speech detection thresholds, and more importantly, had more signs and higher pragmatic skills than the rest of the participants. It is possible that because Participants 2 and 5 had higher language skills in NSL, these translated into oral expressive skills during the intervention. In addition, it seems that oral expressive skills may be dependent on the auditory skills children have. When comparing the results of the current study with the auditory outcomes children showed during the intervention (de Diego-Lázaro & Restrepo, in press), we observed more expressive skills in participants with higher auditory skills except for Participant 1. Participants 2 and 4, who showed increases in auditory detection, discrimination, and identification during the intervention, also increased their production of word imitations. Participant 5 showed increases in all auditory skills, including comprehension, and he communicated using more spontaneous words than imitations. Participant 5 showed the highest MLU among all participants.
The auditory intervention helped children develop some basic expressive vocabulary skills. While it is unknown if more intervention would have helped children develop basic syntax skills, it seems that vocabulary skills are easier to acquire than syntax skills. This is in line with previous studies that have suggested that different language domains may show different critical periods (e.g., Friedmann & Rusou, 2015; Meisel et al., 2013), and that syntax abilities may be particularly challenging to develop after the sensitive period (Friedmann & Szterman, 2005, 2011) Moreover, vocabulary development is a lifelong skill that does not stop developing in typical hearing individuals.
Limitations and Future Directions
This study did not assess children’s abilities in NSL. Teachers and caregivers completed a questionnaire to broadly quantify the number of signs children had in NSL, but no direct language measure was taken (i.e., vocabulary test, language sample, etc.). Future studies should consider language abilities in all languages children know to assess cross-linguistic effects. In addition, the auditory intervention utilized in this study followed a natural approach. The local provider did activities with the children geared to increase their exposure to Spanish, but children were not required to practice or use the targeted words expressively. Future studies should consider including speech intervention strategies to observe if expressive practice enhances word learning.
The teletraining program described in this study could be used to train local providers in developing countries where speech-language pathology studies are still emerging. However, it is important to consider that we did not evaluate the effectiveness of the teletraining program by assessing the local provider’s use of strategies before and after the training. Future studies should formally assess the use of strategies during teletraining programs to inform about what strategies are more effective. In addition, future studies may include quantitative and qualitative measures to show providers’ understanding of concepts and the learning process, particularly when there is a cultural mismatch between the strategies taught (Western culture) and the learner (Nicaraguan culture, in this case).
In conclusion, two of the five participants showed increases in their oral expressive skills during the auditory intervention. It is possible that children’s auditory skills and abilities in NSL helped them increase their use of spontaneous words and imitations in Spanish. Other children, however, may need more explicit intervention to develop oral expressive skills. Finally, teletraining seems to be an effective way to train local providers in developing countries when in-person training is not an option.
Footnotes
Acknowledgements
We want to thank undergraduate students Sarahi Pérez and Nathalia González for transcribing, coding, and helping with reliability in this study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
