Share this article on:

Phantom Stimulation Increases Low-Frequency Cues in CIs

Jiam, Nicole T.; Limb, Charles J. MD

doi: 10.1097/01.HJ.0000479419.67688.d4
Journal Club

Ms. Jiam is a research specialist and a third-year medical student at the Johns Hopkins University School of Medicine.

Dr. Limb is the Francis A. Sooy Professor of Otolaryngology–Head and Neck Surgery and the Chief of the Division of Otology, Neurotology, and Skull Base Surgery at University of California, San Francisco (UCSF). He is also the Director of the Douglas Grant Cochlear Implant Center at UCSF.

Low-frequency sound information may improve music listening outcomes for cochlear implant users, a recent study found.

Over the past 50 years, cochlear implants have made remarkable gains in development and efficacy, particularly with speech optimization. One of the biggest hurdles cochlear implant users still face, though, is music perception and enjoyment. Although cochlear implants are not musically optimized, emerging processing strategies have the potential to deliver sound information that is best appreciated in musical contexts rather than language settings. A new study attempted to determine whether partial bipolar stimulation of low-frequency channels produces benefits for music listening by expanding the frequency range typically available for cochlear implant users. Phantom stimulation shifts the electrical field so insertion depth can be virtually increased by 0.5 to 2 electrodes. Most cochlear implants are able to process frequencies of 300 Hz and higher. Among major cochlear implant companies, Advanced Bionics cochlear implant processors receive sound information above 250 Hz; Cochlear Corporation devices have a low-frequency cutoff of 188 Hz; MED-EL frequency ranges are programmable to a low cutoff of 70 Hz. In the past decade, there has been growing evidence that low-frequency hearing may improve cochlear implant outcomes through means of speech perception (Gifford. Ear Hear 2013;34[4]:413-425; Dorman. Audiol Neurootol 2014;19[4]:234-238; Dorman. Ear Hear 2005;26[4]:371-380; Turner. Hear Res 2008;242[1-2]:164-171. The authors of the study present a new sound processing strategy called Phantom stimulation to transmit low-frequency information by means of a virtual channel.

Design and Evaluation of a Cochlear Implant Strategy Based on a “Phantom” Channel

Nogueira W, Litvak LM, Saoji AA, Büchner A

PLoS One


Monopolar electrode coupling is the common means to deliver cochlear implant electrical stimulation to neurons. When current flows from the primary intracochlear electrode to a remote extracochlear ground contact, an electrical field is formed. These electrical fields can be manipulated by applying reverse phase compensating currents to neighboring electrode contacts (Wilson. Cochlear Implants: Audiological Foundations. San Diego: Singular Publishing Group, 1993.) This technique has since evolved to biphasic pulses in partial bipolar mode. Aniket Saoji and Leonid Livtak called this technique “Phantom electrode stimulation” (Saoj. Ear Hear 2010;31[5]:693-701.

Put simply, when a partial return current is introduced in the electrode adjacent to the most apical electrode, the spread of electrical excitation toward the basal end of the cochlea is reduced and the electrical field shifts toward the apex.

The Phantom processing strategy adds an additional low-frequency channel to the Advanced Bionics HiRes Fidelity 120 (F120) sound processing. This channel was virtually created using partial bipolar stimulation; a primary stimulating current was delivered from electrode contact 1 and a smaller compensating current with opposite phase was discharged by the adjacent electrode contact 2. An apical shift arose from this electrical interaction and provided a lower pitch sensation than that of electrode contact 1 alone. Sigma (σ) is the ratio between the compensating current and the primary current. This value was set during the fitting session, and the spectral bandwidth associated with this channel was large. With regard to stimulation cycles, channel pairs were stimulated sequentially to reduce channel interaction; the order was strategically selected to maximize the distance between stimulation pairs. Partial bipolar stimulation requires a larger charge per phase to produce an equivalent volume to monopolar stimulation. Therefore, the Phantom phase duration was extended to be six times longer than that of the other F120 electrode contacts to reduce the risk of stimulating at greater current levels.

Back to Top | Article Outline


Twelve adult, postlingually deaf, German-speaking cochlear implant users were enrolled in this study. All participants had an Advanced Bionics Clarion CII or HiRes 90k cochlear implant and used the F120 strategy. Users were eligible only if they had good performance with their cochlear implant (near 100% speech intelligibility on the Hochmair Schulz Moser [HSM] sentence test without background noise) to allow for meaningful subjective feedback when using the Phantom coding strategy. Evaluation of the Phantom and F120 strategies was conducted over two sessions; in the first session, impedances were measured for all 16 electrode contacts using a phase duration of 32.32 msec/phase. Impedance values were used to set the upper limit of current (µAmperes) necessary for stimulating individual electrodes. The Phantom strategy was fitted based on the F120 clinical map. Participants were asked to identify the lowest pitch sensation from two stimuli (monopolar vs. Phantom) four times in a blinded, randomized manner. When the user perceived the Phantom stimulus as lower in pitch 100 percent of the time, the corresponding σ value was assigned to the participant. One participant was excluded from the study because the subject was unable to perceive the Phantom stimulus as lower in pitch than the monopolar electrode 1 stimulus, regardless of the σ value. When the Phantom strategy was switched on, participants were asked about sound balance, and the µAmpere level was accordingly adjusted. Speech tests using the HSM sentence test were conducted with both strategies (Hochmair-Desoyer. Am J Otol 1997;18[suppl 6]:S83.

For music perception assessment, cochlear implant users were presented with 20 seconds to 30 seconds of different musical pieces and were asked about their impression of these musical pieces when using Phantom and F120 strategies. For the first session, the musical pieces were “Orchestersuite Nr. 2 H-Moll” by Johann Sebastian Bach and “Serendipity” by Tal Wilkenfeld. Each participants answered a music questionnaire based on a 15-step Likert scale. The questions were: 1) How easy is it to follow the music? 2) How natural does the music sound? 3) How good/natural is the tonal balance of the music? 4) What is the overall impression of the music? and 5) What program do you prefer to use to listen to music? Participants were asked to use the Phantom strategy for the next four weeks before returning for the second session. At the second session, two HSM lists were presented with both the Phantom and F120 strategies. For the musical pieces, participants listened to the same two pieces in the first session and also “You Get What You Give” by the New Radicals.

After one month of Phantom channel use, participants demonstrated significant improvement in speech intelligibility and preference for Phantom in music listening. Most of them reported a preference for the bass-like sound produced by the Phantom during the first session's music perception test. At the second session, about half (n = 6) of the participants showed a strong preference for Phantom and wanted to continue using this strategy as their main program. The other half (n = 5) were either neutral or dissatisfied. The major complaint was that the Phantom channel was too loud. When the µAmpere levels were reduced for these users, they reported an improved sound quality.

There was no significant difference in the HSM sentence test between F120 and Phantom in both sessions. However, there was a trend (p = 0.035) of better performance on Phantom (48.07%) with respect to F120 (36.96%) after four weeks of use. There was a significant improvement in speech intelligibility performance with Phantom between the first and second sessions (40.56% vs. 48.07%, p = 0.008).

With regard to music perception, there was no significant difference between F120 and Phantom for all components of the music questionnaire except for perceived sound balance (p = 0.037). Among listeners, the F120 strategy was reported to be too high pitched, whereas the Phantom strategy provided a more neutral tonal balance. Overall, cochlear implant users preferred to listen to music using the Phantom strategy (n = 10) over the F120 strategy (n = 1).

Back to Top | Article Outline


This study demonstrated a novel method of expanding the low-frequency range by introduction of a virtual channel. Using partial bipolar stimulation, electrical fields were manipulated toward the apical region of the cochlea; as a result, sound information conveyed temporal information beyond the most apical electrode contact. Research in the field of combined electrical and acoustic hearing suggests that low-frequency information can improve speech perception and sound quality (Brown. J Acoust Soc Am 2009;125[3]:1658-1665.

By incorporating Phantom stimulation in the HiFocus 1j cochlear implant, insertion depth is virtually increased by 0.5 to 2 electrodes, or 0.5 to 2 mm of additional insertional depth.

The authors noted that part of the Phantom fitting included reducing the stimulation rate to reduce the amount of current needed to maintain volume. Although there seems to be no overall effect on performance, Waldo Nogueira and colleagues recommend a follow-up study to investigate other alternative settings available with Phantom. Limitations in the study design also arose from the small sample size and the short time interval designated for acclimation to Phantom. The clinical benefits from low-frequency cues may have been clearer if the study were sufficiently powered.

The findings from this study demonstrate contextual preference and improved speech intelligibility after one month of Phantom stimulation. In the Phantom processing strategy, the additional channel transmits an additional bandwidth of two octaves, and the expanded frequency range may provide a more balanced sound quality. There may be positive language benefits as a result of low-frequency hearing. For most users, speech understanding is good to excellent. Conversely, music represents the most challenging acoustic stimulus for cochlear implant users. By report, implant users rank music as the most important acoustic stimulus after understanding speech (Gfeller. J Am Acad Audiol 2000;11[7]:390-406.

Therefore, improved music perception has the potential to improve quality of life for implantees, not only by increasing music enjoyment, but also by enhancing overall hearing and speech understanding in noisy environments. Further research in this field of work is necessary, as the benefits of studying music extend to improving cochlear implant sound perception for the most challenging auditory stimuli in the world.

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.