Abstract
The study attempted to assess the closed set word identification abilities in children with cochlear implant (CI) with a picture identification test developed in Tamil. The test was validated on typically developing children with normal hearing (NH) to build a reference for comparison. Participants for the study included 205 children with NH between the age range of 3–6 years and 45 children with cochlear implant within the age range of 3–11 years. The picture identification test was developed using bisyllabic words with corresponding pictures in Tamil. Two lists were created with 25 words each and administered to the children with NH and CI. The scores of both the groups were analysed. The results indicated that the mean scores improved as age increased for children in the NH group. Also, there was no significant difference in performance between the two-word lists. Significant difference in scores was noted between the CI and the NH group (p < 0.01). However, the mean scores in the CI group increased as the experience with CI increased. The picture identification test in Tamil is deemed appropriate to elicit closed set word identification responses for children with CI between the age of 3–6 years. The test will provide supplemental information for mapping and to plan habilitation for children with CI.
Keywords: Picture identification test, Closed-set word identification, Typically developing children, Cochlear implant
Introduction
Cochlear implantation has become a standard choice of health care intervention for children with profound hearing loss [1]. In India, more than 100,000 babies are born with hearing impairment (HI) every year, and the disability of hearing has a higher prevalence rate among young children below four years of age [2]. Several attempts were made to implement newborn hearing screening programs to bridge the gap between identification and intervention [3]. Considering the colossal impact on the economy, social, and productive life due to disease burden in individuals with hearing loss. The Indian government had initiated funded cochlear implant (CI) programs in many states to address the issue . In 2013, the Government of Tamilnadu started the Chief Minister’s comprehensive Health Insurance scheme (TNCMCHIS). Many children below the age of six with severe to profound sensorineural hearing loss received CI under this scheme [4].
With the increasing numbers of CIs in the state, systematic documentation of program outcomes becomes compelling. The only method to approach it, is by using valid objective measures to document outcomes keeping the global perspective of childhood outcomes in mind. World over many test batteries of speech perception is used to document and to monitor progress in CI recipients . The evaluation of speech perception reflects the interaction of acoustic–phonetic identification, the relation with lexical access, and memory [5]. Test batteries that document the hierarchical development of speech perception in children are available in English, and non-English language like Mandarin Chinese [6]. The test battery includes both indirect measures such as the infant–toddler meaningful auditory integration scale (ITMAIS) and Meaningful Auditory integration Scale (MAIS), which are structured parent questionnaires for evaluating the perceptual skills of young children with CI. Early speech perception test (ESP) directly measures the first evidence of closed set speech perception in children with hearing impairment [7]. Paediatric speech intelligibility testing (PSI) is a closed set recognition task for children between 3 to 6 years [5, 8]. The test consists of monosyllables and sentences in quiet and competing noise. The sentences were presented at different signal to noise ratios. The response task is to point to the picture when the sentence was heard. Lexical Neighbourhood Test (LNT) is a word recognition test with two lists of 50 monosyllabic and Multisyllable Lexical Neighbourhood Test (MLNT) consisting of 24 items. The LNT and MLNT were developed with lexically easy and difficult words on the list based on Logan’s analysis of lexical properties [9]. Hearing in noise test for children (HINT-C) is a 10-sentence list for children with normal hearing and hearing loss between the ages of 6 to 12 years [10].
To monitor progress of children with CI, availability of speech perception tests in the native language is a prerequisite. In Tamil, the availability of validated speech perception test is scarce. Many speech recognition materials have been developed in Tamil, but no tool has been validated for the CI users. Several attempts were made to develop ad -hoc materials that can be used clinically for assessing children [11]. Due to this reason, the progress of children with CI have been largely documented using observation methods such as questionnaires and parental perception. It is well documented that the development of speech perception in children with CI depend on many variables such as age of implantation, CI experience, training, device characteristics and family characteristics [6]. However, it is clinically important to use validated test materials to measure the outcome of a device and plan rehabilitation.
Therefore, the current study attempts to develop a closed set picture identification test and validate the tool with CI users between the ages of 3 to 6 years. The current test material was developed on the operating principle of picture identification test for children with hearing impairment [12]. The main purpose of developing this material is to use it for children who have restricted vocabulary. The response criterion of picture pointing was considered keeping in mind that children with HI have issues in speech production that will influence evaluation. Such as closed-set test for children as young as three years is unavailable in Tamil, considering the number of children being implanted in Tamilnadu, this test will help to document the closed- set identification responses. Children who received CI in Tamilnadu come from various regions of the state, primarily speaking the local dialect. However, much consideration was given during material developed to overcome the dialectal challenges. The periodic assessment using this test will provide the necessary information that will help plan the next step in rehabilitation for any child with CI.
The first section of the study focused on developing a test material that consisted of word lists and pictures. The material was administered on typical children who were deemed to have no developmental disorders between the ages of 3 to 6 years. The data from typical children would serve as a reference for comparing the closed set task with children who use CI. The tool will be further used on children with CI to evaluate the performances and the document their responses.
The primary goal of the study was to validate a picture identification test that can be specifically used for children with CI. Further, variables such the age of implantation and duration of implant use were also examined in the study with regard to the performance with the test tool.
Materials and Methods
The present study was a cross sectional study design and approved by the institutional ethics committee (Ref: IEC-NI/15/FEB/45/10) of Sri Ramachandra University.
The study was conducted in two phases. As mentioned, the first part of the study was primarily to develop the stimuli in Tamil. The material was validated on typically developing children as a norm reference data across the ages and the performances were compared to children with CI.
Selection of stimuli: Picturable words were selected from Tamil school textbooks, story books and from day-to-day conversation. The criteria for selection were that only bisyllabic pictorial words with consonant vowel consonant vowel combination (CVCV) within the vocabulary of the age group under study were considered. Culturally relevant pictures were commercially purchased from online website. To ensure familiarity of the stimuli for children from different regions of the state, common words were selected to avoid dialectal influence. A total of 85 picturable bisyllabic words were shortlisted. The list of words and images was validated for content by four experts, two parents and two schoolteachers who directly work with the children of the age group (3 to 6 years) mentioned in the study. The appropriateness of content and usage of vocabulary for the target population was checked. Words that were unsuitable for the target population were excluded from the list. The stimuli list of 85 words were presented to the children to check for familiarity.
To check the familiarity with the typically developing children with normal hearing (NH), five children in the younger age group (3.1 to 4 years) were tested. The investigator instructed to point to the correct picture whenever the word was heard. To eliminate the chance factor, the investigator also pointed to the picture and asked the child to name it. It was noted, if the picturable word was familiar or unfamiliar to the child. Unfamiliar pictures and the corresponding words were removed from the final list. The final list of 50 picturable bisyllabic words was randomly split into two lists, list A and B with 25 words in each list. Both lists were both phonetically and phonemically balanced based on the frequency of occurrence of phonemes in Tamil language [13]. The template for the picture pointing task consisted of four pictures that were displayed on a PowerPoint slide and displayed on Windows surf 3 tablet with a touch screen. The bisyllabic words stimuli were spoken and recorded by a native female Tamil speaker in a sound recording studio using professional voice recording and Nuendo software (version 6.0). The material was recorded with 44 kHz sampling rate with 16-bit resolution in a sound-treated room. The recorded stimuli were normalized with 1 kHz tone to avoid intensity variations with an inter-stimulus interval of 10 s. The stimuli were subjected to goodness test to ensure 100% recognition of the recorded stimuli by an individual with normal hearing.
The second section of the study focused on the administration of picture identification test in Tamil on children with normal hearing (NH) and children with CI. Two groups were included in the study. One group of participants included N = 205 (M:118; F:87) children with normal hearing (NH) with mean chronological age of testing was 4.8 years (3 to 6 years). Purposive sampling was carried out, and all participants included in the study were native Tamil speakers and were recruited from schools of both urban and rural districts of Tamilnadu. Normal hearing sensitivity was confirmed using air condition thresholds and the pure tone average was within 15dBHL. Otoscopic ear examination and teachers report on poor academic performance were also considered to rule out any history of hearing related issues.
The other group of participants were children with CI, N = 45 (M:26; F:19), mean chronological age of testing was 7.1 years (3–11 years). The mean age of implantation was 3.1 years (1–6 years) and the mean cochlear implant experience was 47 months (3.4- 4.3 years). The participants were Tamil speaking children using CI device (Advanced Bionics-24, Medel-9, and Nuclues-12). All participants received CI at Sri Ramachandra Medical Centre and had attended habilitation for one year. Children with cochlear anomalies and other behavioral issues were not included in the study. The participants within the CI group were grouped into three groups based on the age of implantation (Group 1: 1–2 years; Group 2: 2.1–4 years; Group 3: 4.1–5 years) for analysis. All participants had a stable map and aided thresholds were within 35dBHL. The children used the default speech processing strategy that was recommended by the implant audiologist .
Procedure for Testing
A similar testing procedure was followed for both the groups of typically developing children and children with CI. The testing was carried out in a two-room setup according to the ANSI standard (2003), with the child was seated facing the examiner. The stimulus was presented one by one at a presentation level of 40dBHL via flute Inventis clinical audiometer and loudspeakers placed at one meter and 00 azimuth from the child. Each child was tested individually and instructed to respond by pointing or touching the picture displayed on the Samsung tablet when the stimuli was heard. A trial was provided before the commencement of the actual test. Both lists were administered, and the average testing time was about 15 to 20 min in children with NH whereas children with CI took about 25 to 30 min. The testing was completed with appropriate breaks in between testing to prevent fatigue in the child. The responses were recorded and scored. A correct response was given a score of ‘1’ and incorrect response was scored as ‘0’. Test re-test reliability was performed on 10% of the children with normal hearing and CI.
Statistical Analysis
Descriptive statistics was calculated for the children with NH and CI. Normality was assessed using Shapiro Wilk’s test and results indicated normal distribution of data for both the NH and CI group. Appropriate statistics was used for both the groups. A repeated measure of Analysis of Variance was used to calculate statistical significance for the typically developing children. Paired t- test was used to find the difference in performance between the two lists. For the CI group Kruskal- Wallis test was used to compare the performance across age groups. To compare the performance between the CI and NH group Independent t-test was used.
Results
The performances of typically developing children with normal hearing (NH) based on age was analyzed to establish a normative range for performance. Children with CI were examined based on the age of implantation.
The mean scores of typically developing children with NH (N = 205) for list A and list B were above 95% across the three age groups 3.1 to 4 years, 4.1 to 5 years, and 4.1 to 6 years in both the lists. The effect of age on the scores was statistically analysed, the results indicated the older age group (5.1 to 6 years) performed better in both the lists and the average score was 98% (Table 1). The test was able to assess the closed-set speech perception abilities for children as young as three years of age. A one-way repeated measure of variance analysis indicated that there was a significant difference in performance across the three age groups for list A [F (2,202) = 12.837, p < 0.001] and list B [ F (2,202 = 11.898, p < 0.001] for the test.
Table 1.
Mean and standard deviation of scores of typically developing children with NH using List A & List B of picture identification test in Tamil (PIT) for the three age groups
| Age (n = 205) |
List A Mean (± SD) |
List B Mean (± SD) |
|---|---|---|
| 3.1–4 years | 95.49 (4.6) | 95.19 (4.54) |
| 4.1–5 years | 98.48 (4.46) | 95.63 (4.57) |
| 5.1–6 years | 98.31 (3.378) | 98.08 (3.41) |
The effect of the two wordlists on performance scores were examined and the results indicated that there was no significant difference seen in scores in children with NH (p > 0.01). The results showed that both the word lists were similar and can elicit equal scores in children with NH.
The effect of age of implantation and performance was examined. It was observed that within the CI group the mean raw scores of the children gradually decreased as the age of implantation increased. The overall test scores indicated that children in the CI who were implanted between the age of 1–2 years demonstrated higher performances when compared to the other two groups. Kruskal- Wallis test was conducted to find if there was a significant difference among the groups. However, the results showed no difference in performance based on age of implantation (p > 0.0001) (Table 2).
Table 2.
Mean and standard deviation of performance of children with CI using List A & List B of picture identification test in Tamil (PIT) across the three age groups
| Age of implantation (n = 45) |
List A Mean (± SD) |
List B Mean (± SD) |
|---|---|---|
| 1–2 years | 90.13 (6.90) | 92.13 (5.78) |
| 2.1–4 years | 80.7 (16.71) | 79 (17.35) |
| 4.1–5 years | 73.2 (23.70) | 73 (25.97) |
Independent t-test was used to analyse the performance with the test between children with NH and the CI group. The results indicated that there was significant difference in performance between both the groups (p < 0.01). Also, there was difference in performance between both the groups (CI and NH groups) across the two lists, list A (t (11.12) = 242, p < 0.001) and list B (t (10.40) = 242, p < 0.001).
Discussion
India has a sizeable population of children and adults using CI, and it will be on an upward trend in the future. Validated test materials in the native language are the pressing need to plan and monitor the (re)habilitation process in the CI program programs. In the context of the present study, the validation of a closed set picture identification test for evaluating children's hierarchical performance with CI will be a clinically objective tool. The picture identification test was administered on typically developing children with NH to establish a reference to compare the performance of children with CI [14]. The results of NH group suggest children as young as three years can perform the test. The scores of NH group reached ceiling, which indicate that the test was age-appropriate [5]. The results suggested that typically developing children have significant differences in performances as age advances [8]. The material developed in the current study will serve as a data pool of validated stimuli that is phonetically and phonemically balanced in Tamil. It was noted that the performance of children with NH across both the test lists were equal. Therefore, it is established that both the lists are equal and can provide equal scores.
In the current study all the participants could perform the test, which specified that the dialectal difference did not impact the stimuli perception. The stimuli could be used to develop tests for children with other sensory devices.
In the assumption that the children with hearing impairment using CI will perform the test similarly, the test was administered to CI population. The children required familiarity and orientation, additionally to complete the test [8, 15]. The children were able to follow instruction, the children did not find it challenging and all children were able to complete it. The children were divided in three groups based on the age of implantation. It was noted that the performance of children with CI increased with the duration of implant usage. The performance on the test as a function of age was examined.
The current study results indicated that children who were implanted early (1–2 years) showed a significant increase in performance for the picture identification test compared to the other two groups [15]. Children implanted at a younger age have a reduced auditory deprivation duration and are provided with appropriate auditory stimulation to develop speech and language during the sensitive periods [16]. Therefore, children implanted younger have rapid growth trajectories compared to older children implanted later than two years [6, 17]. Also, children implanted in the age range between 1–2 years have a longer duration of implant experiences, and that served as an advantage to improve the performance of children with the closed set test. Similar findings were also reiterated that fact that as the age of implantation increased, the scores of the picture identification test also decreased correspondingly. [1]
The normative data for the typically developing children served as a standard to compare the CI group for the Tamil speaking population. The performance of the CI group has always demonstrated a great deal of variability in speech recognition. However, the key findings in the present study indicated that some children the CI group also performed as well as their hearing peers.
Nevertheless, it is well-established fact that the speech performance scores of children with CI depend on various other factors. The inherent limitation for the study may be that the performance of children with device failure, hardware issues, and intermittent usage [18] may not be reflected. The study attempted to develop a validated closed set material in Tamil of picture identification test for children with CI and for as young as 3 years of age. The data from typically developing children can serve as a base for age referenced norms in children with hearing impairment using a CI device.
Conclusion
The picture identification test was developed as part of the hierarchy of speech perception test for preschool children in Tamil for children with CI between the ages of 3–6 years. For this purpose, the test material was validated on typically developing children to obtain the age referenced norms. The results have demonstrated that the test is an age appropriate tool for typically developing children as well as for children with hearing impairment who use CI. As the number of children receiving implantation in Tamilnadu is increasing with the government schemes and self-funding CI surgeries. The current test could serve as an objective measure to evaluate and document the closed set identification skill in CI recipients. The evidence from the study indicate that the test can be used as part of clinical documentation of speech perception skills used in the test battery for assessing children with CI. As future directive it would be worthwhile evaluating the word recognition skills of children with CI using in the presence of noise using this test material.
Declarations
Funding
No funding was obtained for the study.
Declarations
Conflict of interests
There was no conflict of interest among the authors.
Ethical Standards
The study was conducting after approval for the institutional ethics committee of SRIHER (Ref: IEC-NI/15/FEB/45/10).
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Dawson Gladys Prathiba, Email: prathiba.dawson@sriramachandra.edu.in.
A. Ravikumar Arunachalam, Email: arunachalamravikumar1954@gmail.com.
Prakash Boominathan, Email: prakash_boominathan@sriramachandra.edu.in.
References
- 1.Govaerts PJ, Beukelaer C De, Daemers K, Ceulaer G De, Yperman M, Somers T et al (2002) Outcome of Cochlear Implantation at Different Ages from 0 to 6 Years. (January 1994):885–90 [DOI] [PubMed]
- 2.Galhotra A, Sahu P (2019) Challenges and solutions in implementing hearing screening program in India. 299–302 [DOI] [PMC free article] [PubMed]
- 3.Zheng Y, Soli SD, Meng Z, Tao Y, Wang K, Xu K et al (2010) Assessment of Mandarin-speaking pediatric cochlear implant recipients with the Mandarin Early Speech Perception (MESP) test. Int J Pediatr Otorhinolaryngol [Internet] 74(8):920–5. Available from: http://www.sciencedirect.com/science/article/pii/S0165587610002247 [DOI] [PubMed]
- 4.Arulalan SL, Vadivu S, Dharmarajan S, Dham R (2020) International journal of pediatric otorhinolaryngology cochlear implantation in different socioeconomic groups—bursting the myth. Int J Pediatr Otorhinolaryngol [Internet] 136(1):110156. 10.1016/j.ijporl.2020.110156 [DOI] [PubMed]
- 5.Cokely CG, Tobey A, Quittner AL, Fink E, Wang N, Niparko JK, et al. Speech recognition at 1-year follow-up in the childhood development after cochlear implantation study. Meth Prelim Find. 2006;90057:259–68. doi: 10.1159/000093302. [DOI] [PubMed] [Google Scholar]
- 6.Chen Y, Wong LLN (2017) Speech perception in Mandarin-speaking children with cochlear implants: a systematic review. Int J Audiol [Internet] 56(sup2):S7–16. 10.1080/14992027.2017.1300694 [DOI] [PubMed]
- 7.Geers AE, Jean SM (2012) Early speech perception. Central institute for the deaf, St. Louis, Missouri. print. Early speech perception. St. Louis, Missouri: Central Institute for the Deaf
- 8.Zheng Y, Soli SD, Wang K, Meng J. Development of the Mandarin pediatric speech intelligibility ( MPSI ) test. Int J Audiol. 2009;48:718–728. doi: 10.1080/14992020902902658. [DOI] [PubMed] [Google Scholar]
- 9.Martinez A, Kirk KI, Eisenberg LS, Martinez AS, Hay-mccutcheon M (1999) The lexical neighborhood test: test-retest reliability and inter-list research on spoken language processing. In: The lexical neighborhood test: test-retest reliability. Indiana University (July 2015)
- 10.Nilsson M, Soli SD, Sullivan J (1993) Development of the hearing. In: Noise test for the measurement of speech reception thresholds in quiet and in noise the BKB sentence materials, pp 1085–99 [DOI] [PubMed]
- 11.Boominathan P. Picture Identification test for children with hearing imapirement in Tamil. India: Mysore University; 1999. [Google Scholar]
- 12.Ross M, Lerman JAY. A picture identification hearing-impaired. JSHR. 1970;13(1):44–53. doi: 10.1044/jshr.1301.44. [DOI] [PubMed] [Google Scholar]
- 13.Ramakrishnan BS, Nair KK, Chiplunkar VN, Atal BS, Ramachandran V, Subramanian, R (1962) Some aspects of relative efficiencies of Indian Language. Department of Electrical Communication Engineering, Indian Institute of Sciences, pp 1–88
- 14.Houston DM, Pisoni DB, Kirk KI, Ying EA, Miyamoto RT. Speech perception skills of deaf infants following cochlear implantation: a first report. Int J Pediat Otorhinol. 2003;5876(479):495. doi: 10.1016/s0165-5876(03)00005-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Anderson I, Martin J, Costa A, Jamieson L, Bailey E, Plant G, et al. Validation of the Common Objects Token (COT) Test for children with cochlear implants. Deaf Educ Int. 2005;7(3):154–170. doi: 10.1179/146431505790560347. [DOI] [Google Scholar]
- 16.Vlastarakos PV, Proikas K, Papacharalampous G, Exadaktylou I, Mochloulis G, Nikolopoulos TP. International journal of pediatric otorhinolaryngology cochlear implantation under the first year of age—the outcomes. Critic Syst Rev Meta-Anal. 2010;74:119–126. doi: 10.1016/j.ijporl.2009.10.004. [DOI] [PubMed] [Google Scholar]
- 17.Svirsky MA, Teoh S. Development of language and speech perception in congenitally. Profoundly Deaf Child Fun Age Cochlear. 2004;5200:224–233. doi: 10.1159/000078392. [DOI] [PubMed] [Google Scholar]
- 18.Sarant J, Naz S. Cochlear implants in children: a review. Hear Loss. 2012;122:356–362. [Google Scholar]