Mary Joe Osberger

Independent evaluation of the speech perception abilities of children with the nucleus 22-channel cochlear implant system

The performance of 28 children with the Nucleus multichannel cochlear implant, who had used the device an average of 1.7 yr, was examined on a battery of speech perception measures. All children demonstrated better speech perception skills with the implant than they had in the preimplant condition with hearing aids. With the Nucleus implant, 61% of the children demonstrated some open-set speech recognition and another 14% demonstrated closed-set speech recognition. Scores on the tests were corrected for guessing and a hierarchy of test difficulty was developed. The results revealed systematic differences in performance as a function of perception task and test format. The results of regression analyses, which were performed to identify predictors of success, showed that communication mode made a significant unique contribution to the variance in performance among subjects on an open-set word recognition test. When the scores of the children who used oral or total communication were compared on the full battery of tests, however, there were few significant group differences.

Variables affecting implant performance in children

This study examined the variables that contribute to the large individual differences in the speech perception skills of children with the Nucleus multichannel cochlear implant. Sixty‐one children were tested on four measures of speech perception: two tests of closed‐set word recognition, one test of open‐set recognition of phrases, and one open‐set monosyllabic word test, scored on the basis of the percentage of phonemes as well as words identified correctly. The results of a series of multiple regression analyses revealed that the variables of processor type, duration of deafness, communication mode, age at onset of deafness, length of implant use, and age implanted accounted for roughly 35% of the variance on two tests of closed‐set word recognition, and 40% of the variance on measures that assessed recognition of words or phomenes in an open set. Length of implant use accounted for the most variance on all of the speech perception measures.

Assessing Cochlear Implant Benefit in Very Young Children

3. Tyler RS, Fryauf-Bertschy H, Kelsay DMR, Gantz BJ, Woodworth GP, Parkinson A. Speech perception by prelingually deaf children using cochlear implants. Otolaryngol Head Neck Surg 1997;117:1807. 4. Balkany T, Hodges AV, Luntz M. Update on cochlear implantation. Otolaryngol Clinic North Am 1996;29:277-89. 5. Miyamoto R, Osberger M, Todd S, et al. Variables affecting implant performance in children. Laryngoscope 1994;104:1120-4. 6. Harrison R, Nedzelski J, Picton N, et al. The Paediatric Cochlear Implant Program at the Hospital for Sick Children, Toronto. J Otolaryngol 1997;26:180-7. 7. De Jong A, Nedzelski J, Papsin B. Surgical outcome measures of pédiatrie cochlear implantation. J Otolaryngol 1998;27:26-30. 8. Dawson PW, Blarney PJ, Rowland LC, et al. Cochlear implants in children, adolescents, and prelinguistically deafened adults: speech perception. J Speech Hear Res 1992;35:401-17.

Using current steering to increase spectral resolution in CII and HiRes 90K users.

Objectives: The HiResolution Bionic Ear has the capability of creating virtual spectral channels using current steering. Through simultaneous delivery of current to pairs of adjacent electrodes, it is hypothesized that the effective locus of stimulation can be steered to sites between the contacts by varying the proportion of current delivered to each electrode of the pair. Thus, theoretically, many intermediate regions of stimulation can be created with fine control over the proportion and amplitude of current delivered to each electrode. This study investigated the number of spectral channels—or different pitches—that could be resolved by adult users of the CII and HiRes 90K cochlear implants when current steering was applied to three pairs of electrodes along the implanted array. Design: Subjects were postlinguistically deafened adults recruited from the general CII and HiRes 90K user populations at 11 participating study sites. After loudness balancing and pitch ranking electrode pairs (2 and 3, 8 and 9, 13 and 14), an adaptive paradigm was used to estimate the number of intermediate pitch percepts that could be heard for each pair when current steering was implemented. Those data were used to estimate the potential number of spectral channels for each electrode pair. Results: Data from 57 implanted ears indicated that the numbers of spectral channels per electrode pair ranged from one (subjects who could not tell the electrodes apart) to 52 (an individual who had 52 different pitch percepts for the midarray pair of electrodes). The average numbers of spectral channels that could be distinguished were 5.4 for the basal electrode pair, 8.7 for the midarray electrode pair, and 7.2 for the apical electrode pair. Assuming that the average numbers of spectral channels for these three electrode pairs were representative of the entire 16-contact array, the potential total numbers of spectral channels could be estimated. For the 57 ears, the number of potential channels ranged from 8 to 466, with an average of 93. Conclusions: The HiResolution Bionic Ear has the ability to steer current through simultaneous stimulation of adjacent electrode contacts. These data show that the majority of subjects perceive additional spectral channels other than those associated with stimulation of the fixed electrodes when current steering is implemented. The results suggest that the average cochlear implant user may have significantly more place-pitch capability than is exploited presently by cochlear implant systems. Current steering will be implemented in a wearable sound-processing strategy that can deliver up to 120 spectral channels to CII and HiRes 90K recipients. The new strategy takes advantage of untapped capabilities of the CII/HiRes 90K implanted electronics and will be implemented through software, with no additional surgery required. It is anticipated that the improved spectral resolution offered by current steering will lead to better speech perception in noise and improved music appreciation.

Cochlear implant candidacy and performance trends in children.

Data from Clarion cochlear implant pediatric clinical trials were examined retrospectively to uncover trends in candidacy and postimplant benefit over time. In particular, age at implantation, educational setting, and communication mode were examined with respect to speech perception performance after implantation. The results showed that 1) age at implantation is decreasing, 2) children in oral education programs obtain more benefit from a cochlear implant than children in total communication programs, 3) children who undergo implantation before 2 years of age show greater benefit than children who undergo implantation between 2 and 3 years of age, 4) more younger children are using oral communication than older children, and 5) more children with good auditory skills before implantation and more residual hearing are undergoing implantation. In sum, in the 11 years since implants have been available to children in the United States, candidacy criteria have evolved and benefit has increased as cochlear implant technology has advanced.

HiResolution and conventional sound processing in the HiResolution bionic ear: using appropriate outcome measures to assess speech recognition ability.

Objective: This study compared speech perception benefits in adults implanted with the HiResolutionTM (HiRes) Bionic Ear who used both conventional and HiRes sound processing. A battery of speech tests was used to determine which formats were most appropriate for documenting the wide range of benefit experienced by cochlear-implant users. Study Design: A repeated-measures design was used to assess postimplantation speech perception in adults who received the HiResolution Bionic Ear in a recent clinical trial. Patients were fit first with conventional strategies and assessed after 3 months of use. Patients were then switched to HiRes sound processing and assessed again after 3 months of use. To assess the immediate effect of HiRes sound processing on speech perception performance, consonant recognition testing was performed in a subset of patients after 3 days of HiRes use and compared with their 3-month performance with conventional processing. Setting: Subjects were implanted and evaluated at 19 cochlear implant programs in the USA and Canada affiliated primarily with tertiary medical centers. Patients: Patients were 51 postlinguistically deafened adults. Main Outcome Measures: Speech perception was assessed using CNC monosyllabic words, CID sentences and HINT sentences in quiet and noise. Consonant recognition testing was also administered to a subset of patients (n = 30) using the Iowa Consonant Test presented in quiet and noise. All patients completed a strategy preference questionnaire after 6 months of device use. Results: Consonant identification in quiet and noise improved significantly after only 3 days of HiRes use. The mean improvement from conventional to HiRes processing was significant on all speech perception tests. The largest differences occurred for the HINT sentences in noise. Ninety-six percent of the patients preferred HiRes to conventional sound processing. Ceiling effects occurred for both sentence tests in quiet. Conclusions: Although most patients improved after switching to HiRes sound processing, the greatest differences were seen in the ‘poor’ performers because ‘good’ performers often reached ceiling performance, especially on tests in quiet. Future evaluations of cochlear-implant benefit should make use of more difficult measures, especially for ‘good’ users. Nonetheless, a range of difficulty must remain in test materials to document benefit in the entire population of implant recipients.

Prelingually deafened children’s performance with the nucleus multichannel cochlear implant

The speech perception abilities of 19 children with onset of deafness before age 3 years was examined after they received the Nucleus multichannel cochlear implant. The children were divided into two groups based on age at onset of deafness: children with congenital deafness (n = 8) and children with onset of deafness after birth but before age 3 (n = 11). There was no statistically significant difference between the scores of the two groups of subjects on 12 of the 13 speech perception tests administered. This finding suggests that children who are born deaf have the potential to derive the same benefit from cochlear implants as do children who have had some exposure to spoken language before the onset of their deafness. Examination of performance in terms of communication mode revealed that prelingually deafened children with implants who used oral communication obtained significantly higher scores on only 2 of the 13 speech perception measures than did children who used total communication. The data suggest that communication mode does not appear to account for large differences in speech perception performance among prelingually deafened children with multichannel cochlear implants.

Effect of Age at Onset of Deafness on Children's Speech Perception Abilities with a Cochlear Implant

The speech perception abilities of 37 children with cochlear implants (single-channel or multichannel) were examined as a function of age at onset of deafness. There was no significant difference in the speech perception abilities of implanted children who were born deaf and those of implanted children who lost their hearing during the first 3 years of life. In contrast, the performance of children whose age at onset of deafness was 5 years or later was significantly better than that of the children with congenital or early-acquired deafness on tests of stress pattern categorization, closed-set word identification, open-set identification of common phrases, and lipreading enhancement.

The effect of timing errors on the intelligibility of deaf children's speech

At the 93rd meeting of the Acoustical Society of America, we presented a preliminary report on a technique to correct deviant timing errors in deaf childrens speech by computer processing of recorded speech samples. Timing modifications were accomplished by direct manipulation of the waveform in which whole pitch periods were deleted from steady‐state portions of speech segments, leaving all other aspects of the speech unchanged. The following six‐stage approximation procedure was used to correct the deviant timing patterns: (1) original, unaltered utterance, (2) correction of pauses only, (3) correction of relative timing (stressed/unstressed ratio), (4) correction of absolute syllable duration, (5) correction of relative timing and pauses, and (6) correction of absolute syllable duration and pauses. This report will present preliminary data on the changes in intelligibility that occurred as a result of the systematic timing corrections Intelligibility measures were obtained by playing the original sent...

Imitative consonant feature production by children with multichannel sensory aids.

Objective: To examine changes over time in consonant feature production by children with profound hearing impairments who used either the Nucleus multichannel cochlear implant or the multichannel vibrotactile aid, Tactaid 7. Design: Imitative consonant productions of children with prelingual deafness were elicited and transcribed at two intervals: 1) before receiving their respective devices (predevice interval), and 2) after an average of 1.5 yr of device use (postdevice interval). The consonant productions were analyzed in terms of the percentage of consonant features (manner, place, and voicing) produced by the child that matched the features of the examiners target. The percentage of features produced correctly was then averaged across repetitions, vowel environments, and participants within each group. Results: At the predevice interval, the cochlear implant and Tactaid 7 participants demonstrated similar imitative consonant production abilities. After an average of 1.5 yr of device use, the cochlear implant participants demonstrated significantly greater gains than did the Tactaid 7 participants for the features of voicing and place of articulation. Although the cochlear implant participants showed a trend towards better production of the consonant manner features, this difference failed to reach significance. Conclusions: The current results suggest that the use of a multichannel sensory aid yields improvements in consonant feature production. Furthermore, use of a cochlear implant appears to promote the production of consonant voicing and place features to a greater degree than does the use of a multichannel tactile aid.