Bilateral cochlear implants controlled by a single speech processor.

OBJECTIVE: This study aimed to assess, in one profoundly hearing impaired subject, potential benefits and limitations in placing bilaterally implanted scala tympani electrode arrays under control of a single speech processor. STUDY DESIGN: All available stimulation sites in both ears were compared in studies of pitch discrimination and pitch ranking, identifying three bilateral pairs capable of supporting interaural comparisons with no perceptible difference in pitch. Using those pairs, the subject's ability to lateralize sound was studied as a function of interaural time delay and interaural amplitude difference. Consonant identification scores were obtained for continuous interleaved sampling processors using various unilateral and bilateral combinations of electrodes. RESULTS: For loudness-matched stimuli composed of 50-msec bursts of 80-microsec/phase pulses at 480 pulses/sec, the subject was able to identify the ear receiving the earlier onset for interaural delays at least as brief as 150 microsec for all three matched pairs. For similar simultaneous stimuli, the subject could identify the ear receiving the louder signal for the smallest deviations from loudness-matched amplitudes available from the implanted electronics. The consonant studies found no evidence that bilateral stimulation per se degrades speech processor performance, even for arbitrary divisions of information between the two ears. Additional contralateral as well as ipsilateral channels were observed to improve speech processor performance. CONCLUSIONS: The ability of this subject to lateralize sounds on the basis of interaural delay or loudness difference, combined with the consonant identification results, supports further use of coordinated binaural stimulation to improve cochlear implant users' ability to understand speech, especially in the presence of competing speech noise.

Cochlear implant studies at Research Triangle Institute and Duke University Medical Center.

Examples from several areas of cochlear implant research are presented, with emphasis on the continuous interleaved sampling (CIS) approach to speech processor design. Within-subject comparisons of such processors with the compressed analog (CA) approach of the clinical Ineraid device are reviewed, and ongoing similar comparisons with the clinical Nucleus spectral peak (SPEAK) strategy are outlined. Correlations between chronic performance levels with clinical CA processors and initial performance levels with CIS, data on further improvements in performance with chronic use of CIS, and instances of substantial benefit from custom fitting of CIS parameters are presented as examples of findings with immediate clinical implications. New studies are described, involving the measurement of intracochlear evoked potentials in response to cochlear implant stimulation, and the integration of such work with computer modeling studies.

Use of maximum length sequence analysis in newborn hearing testing.

The use of maximum length sequence analysis (MLSA) with rapid click rates may be of clinical value in newborn hearing screening because a greater number of individual responses can be signal averaged without adding to test time. To examine the potential clinical value of MLSA in newborn screening, auditory brainstem responses (ABRs) from 50 premature newborns were studied. ABRs were acquired with conventional signal averaging at four stimulus intensity levels (50 dB, 40 dB, 30 dB, and 20 dB nHL) using a click rate of 33.3/sec. These responses were directly compared with the ABRs acquired with MLSA using a rate of 227.3/sec. MLSA and conventional signal averaging yielded similar results with no statistically significant differences in the number of responses detected. Across babies, the overall quality of the tracings slightly favored MLSA, particularly when recording conditions were poor. Though the results of this investigation do not support the use of MLSA as the primary technique in newborn screening in the neonatal intensive care unit (NICU), they do support consideration of the use of MLSA as an alternative technique when the responses obtained with conventional signal averaging are poorly defined.

Application of maximum length sequence analysis to auditory brainstem response testing of premature newborns.

The feasibility of utilizing maximum length sequence analysis (MLSA) with newborns was examined on 40 premature infants residing in a neonatal intensive care unit (NICU). Auditory brainstem responses (ABRs) were initially obtained using a conventional stimulus rate of 33.3/sec. Using MLSA, responses were also recorded using much faster stimulus rates (227.3/sec, 454.5/sec, and 909.1/sec). As expected, faster stimulus rates resulted in longer ABR component wave latencies. Less expected was the clarity of the ABRs that were obtained through MLSA. Well-defined responses were obtained from premature infants at all stimulus rates. The findings suggest that MLSA may have clinical applications to newborn hearing screening.