Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jul 21.
Published in final edited form as: Audiol Neurootol. 2023 Jul 21;28(6):478–487. doi: 10.1159/000531262

Comparison of two place-based mapping procedures on masked sentence recognition as a function of electrode array angular insertion depth and presence of acoustic low-frequency information: A simulation study

Margaret T Dillon 1,2, Emily Buss 1, Alec D Johnson 1, Michael W Canfarotta 1, Brendan P O’Connell 1,3
PMCID: PMC10948008  NIHMSID: NIHMS1970713  PMID: 37482054

Abstract

Introduction:

Cochlear implant (CI) and electric-acoustic stimulation (EAS) users may experience better performance with maps that align the electric filter frequencies to the cochlear place frequencies, known as place-based maps, than with maps that present spectrally shifted information. Individual place-based mapping procedures differ in the frequency content that is aligned to cochlear tonotopicity versus discarded or spectrally shifted. The performance benefit with different place-based maps may vary due to individual differences in angular insertion depth (AID) of the electrode array and whether functional acoustic low-frequency information is available in the implanted ear. The present study compared masked speech recognition with two types of place-based maps as a function of AID and presence of acoustic low-frequency information.

Methods:

Sixty adults with normal hearing listened acutely to CI or EAS simulations of two types of place-based maps for one of three cases of electrode arrays at shallow AIDs. The strict place-based map (Strict-PB) aligned the low- and mid-frequency information to cochlear tonotopicity and discarded information below the frequency associated with the most apical electrode contact. The alternative place-based map (LFshift-PB) aligned the mid-frequency information to cochlear tonotopicity and provided more of the speech spectrum by compressing low-frequency information on the apical electrode contacts (i.e., < 1 kHz). Three actual cases of a 12-channel 24-mm electrode array were simulated by assigning the carrier frequency for an individual channel as the cochlear place frequency of the associated electrode contact. The AID and cochlear place frequency for the most apical electrode contact was 460º and 498 Hz for Case 1, 389º and 728 Hz for Case 2, and 335º and 987 Hz for Case 3, respectively.

Results:

Generally, better performance was observed with the Strict-PB maps for Cases 1 & 2, where mismatches were 2–4 octaves for the most apical channel with the LFshift-PB map. Similar performance was observed between maps for Case 3. For the CI simulations, performance with the Strict-PB map declined with decreases in AID, while performance with the LFshift-PB map remained stable across cases. For the EAS simulations, performance with the Strict-PB map remained stable across cases, while performance with the LFshift-PB map improved with decreases in AID.

Conclusions:

Listeners demonstrated differences with the Strict-PB versus LFshift-PB maps as a function of AID and whether acoustic low-frequency information was available (CI versus EAS). These data support the use of the Strict-PB mapping procedure for AIDs ≥ 335º, though further study including time for acclimatization in CI and EAS users is warranted.

Keywords: cochlear implant, electric-acoustic stimulation, frequency-to-place mismatch, spectral shift, cochlear tonotopicity, filter frequency

Introduction

Cochlear implant (CI) recipients who listen with either CI-alone or electric-acoustic stimulation (EAS) devices may experience better speech recognition with place-based maps than with spectrally shifted maps [Dorman, Loizou, & Rainey, 1997; Shannon, Zeng, & Wygonski, 1998; Fu & Shannon, 1999b; Başkent & Shannon, 2003; Faulkner, Rosen, & Stanton, 2003; Başkent & Shannon, 2007; Li & Fu, 2010; Dillon, 2022; Dillon et al., 2023]. Place-based mapping procedures typically assign some or all of the electric filter frequencies to match the cochlear place frequencies for the electrode contacts to eliminate spectral shifts, also known as frequency-to-place mismatches. Individual place-based mapping procedures differ in the frequency content that is aligned to cochlear tonotopicity versus discarded or spectrally shifted. The performance benefit with different place-based maps may vary due to individual differences in the angular insertion depth (AID) of the electrode array and whether functional acoustic low-frequency information is available in the implanted ear. The present study compared masked speech recognition with two types of place-based maps for participants listening to CI or EAS simulations based on electrode arrays at different AIDs.

In an ongoing study with CI and EAS users, we are investigating the effectiveness of a strict place-based mapping procedure (referred to here as Strict-PB). Our Strict-PB procedure assigns the electric filter frequencies to match the cochlear place frequencies for the contacts in the low- to mid-frequency region (up to at least 3 kHz) and distributes the remaining high-frequency information across the basal contacts. The rationales for aligning the low- and mid-frequency information are: 1) better spectral resolution of low-frequency information may support outcomes in competing maskers [Qin & Oxenham, 2003; Jin & Nelson, 2010], and 2) aligning the critical speech information [e.g., 1–3 kHz; Warren et al., 1995; ANSI, 1999] may support better speech recognition. The rationales for distributing the high-frequency information across basal contacts as opposed to deactivating contacts in the region above the upper filter frequency limit for the device (i.e., 8.5 kHz) are: 1) listeners can tolerate spectral shifts of high-frequency information when the mid-frequency information is aligned [Başkent & Shannon, 2007] , and 2) most high-frequency speech cues are not spectrally discrete (e.g., voiceless fricatives). Low-frequency information below the frequency associated with the most apical contact is discarded to eliminate frequency-to-place mismatches.

A consideration of the Strict-PB procedure is that recipients of electrode arrays at shallow AIDs may experience limited benefit due to the loss of low-frequency information [Fu & Shannon, 1999b; Faulkner et al., 2003; Başkent & Shannon, 2005]. For example, Başkent and Shannon [2005] simulated insertion depths ranging from 28 mm to 7 mm in 4 CI users by deactivating apical contacts. For deeper insertion depths, CI users had better performance with the place-based maps compared to the spectrally shifted maps. Performance declined with the place-based maps as the simulated insertion depth decreased. At approximately 19 mm (center frequency, CF: 1332 Hz), CI users began performing better with the spectrally shifted maps than the place-based maps. These findings demonstrate that discarding low-frequency information, such as with our Strict-PB mapping procedure, may be detrimental for the speech recognition of recipients of electrode arrays at shallow AIDs.

As compared to the Strict-PB procedure, recipients of electrode arrays at shallow AIDs may experience better performance with maps derived from an alternative place-based mapping procedure that provides more low-frequency information, albeit in spectrally shifted fashion (referred to here as LFshift-PB). The LFshift-PB mapping procedure aligns the critical mid-frequency speech information and compresses lower and higher frequency information to provide more of the speech spectrum. Thus, the Strict-PB and LFshift-PB procedures differ in the distribution of low-frequency information. While both approaches respect tonotopicity more than current default mapping procedures by aligning mid-frequency speech information, presenting low-frequency information with intentional spectral shifts (LFshift-PB) may offer an advantage over discarding low-frequency information (Strict-PB) for electrode arrays at shallow AIDs. For CI simulations, we hypothesized that performance with Strict-PB maps might deteriorate with decreases in AID due to the progressive reduction in low-frequency information provided, and that LFshift-PB maps might support better performance in these conditions due to the provision of low-frequency information.

We also hypothesized that the pattern of performance with Strict-PB versus LFshift-PB maps as a function of AID may be different for CI versus EAS users. EAS simulations demonstrate a speech recognition benefit with strict place-based maps as compared to spectrally shifted maps, even when the strict place-based map results in a frequency information gap between the acoustic and electric outputs [Fu, Galvin, & Wang, 2017; Willis et al., 2020; Dillon et al., 2021, 2022]. Acoustic low-frequency hearing may serve as an anchor to cochlear tonotopicity – limiting the ability to benefit from spectrally shifted electric information in the LFshift-PB map. If that is the case, then the benefit associated with Strict-PB maps might extend over a wider range of AIDs for simulations of EAS than for CI.

The present study aimed to 1) determine whether presenting low-frequency information with spectral shifts offers an advantage over discarding low-frequency information when the mid-frequency information is aligned with cochlear place for simulations of electrode arrays at shallow AIDs, and 2) evaluate whether the addition of acoustic low-frequency cues with EAS influences performance differences with the two place-based maps as a function of AID.

Methods

The study-site Biomedical Institutional Review Board (IRB) approved the assessment of speech recognition while listening to vocoded speech (IRB approval #86–0059). Listeners provided written consent and were compensated $15.00 per hour.

Listeners

Sixty young adults (40 female) participated. Listeners were between 18 and 29 years of age, with a mean age of 23 years (SD: 3 years). Hearing sensitivity was assessed behaviorally in either a single-walled or double-walled soundbooth. Listeners detected pure-tone stimuli of ≤ 25 dB HL for octave frequencies .125 to 8 kHz and for 12.5 kHz. Listeners were native speakers of American English with no previous listening experience to vocoded speech.

Stimuli

A 12-channel noise vocoder simulated the electric outputs for the CI and EAS simulations using a bank of bandpass finite impulse response (FIR) filters. Tap arrays were generated using the fir1 function (MATLAB, 2019a). For each filter, the number of taps used to define the magnitude spectrum was selected such that spectral resolution was 20% of the filter bandwidth; synchronous output across filters was accomplished by symmetrically padding the arrays with zeros. The CFs for the bandpass filters were the CFs for each channel derived from either the Strict-PB or LFshift-PB procedure for three cases of electrode arrays at shallow AIDs from our patient population. Figure 1 plots the AID of each contact by the CF for each channel for the Strict-PB and LFshift-PB maps for each case (listed in Supplemental Table 1). The solid black line indicates the estimated cochlear place frequency using a spiral ganglion (SG) frequency-to-place function, as described by Stakhovskaya et al [2007]. The Hilbert envelope was extracted from the output of each filter, low-pass filtered at 300 Hz with a 4th order Butterworth filter, and used to amplitude modulate a corresponding noise band. The noise bands were filtered using the cochlear place frequencies for each case (plotted in Figure 1 and listed in Supplemental Table 1).

Figure 1:

Figure 1:

Open in a new tab

Angular insertion depth of each electrode contact as a function of the channel center frequency (CF) for the Strict-PB and LFshift-PB maps. The solid black line indicates the cochlear place frequency estimated using a spiral ganglion frequency-to-place function, as described by Stakhovskaya et al (2007). Results for the three cases are shown in separate panels.

For the EAS simulations, acoustic low-frequency information was added to the noise vocoder. A FIR filter shaped the acoustic output to simulate aided sound field thresholds of 30, 30, 55, and 65 dB HL at .125, .25, .5, and 1 kHz, respectively. The rationale was to simulate the acoustic low-frequency information available to an EAS user with a moderate hearing loss at .125 and .25 kHz and a severe hearing loss at .5 and 1 kHz. The acoustic hearing was not incorporated into the electric filter assignments of the Strict-PB and LFshift-PB maps, which is in contrast with the current clinical default mapping procedures for EAS devices. For the present study, vocoder settings were consistent for the EAS and CI simulations to facilitate a direct performance comparison with and without acoustic low-frequency cues. This procedure avoided the differences in frequency-to-place mismatches for EAS and CI simulations that would have occurred if the acoustic hearing had been used to define the electric low-frequency filter for the EAS simulations. This procedure resulted in some overlap of the acoustic and electric input frequencies in the EAS-LFshift condition for Case 1 and Case 2 (discussed below).

Simulated Cases of Short Electrode Arrays

Three actual cases of a 12-channel 24-mm electrode array were simulated by assigning the carrier frequency for an individual channel as the cochlear place frequency of the associated contact (plotted in Figure 1 and listed in Supplemental Table 1). The AID and SG cochlear place frequency for the most apical contact was 460º and 498 Hz for Case 1, 389º and 728 Hz for Case 2, and 335º and 987 Hz for Case 3, respectively. These example cases from the study-site clinical population were selected because they are within the range of AIDs reported for recipients of < 25-mm electrode arrays that are used clinically [Landsberger et al., 2015; O’Connell et al., 2016; Canfarotta et al., 2020; Lenarz et al., 2020]. For example, the average AID of the most apical contact are 428º (SD: 34º) for the MED-EL Flex24 array [Canfarotta et al., 2020], 411º (SD: 78º) for the Cochlear CI422 Slim Straight array [O’Connell et al., 2016], and 393º (SD: 62º) for the Advanced Bionics SlimJ array [Lenarz et al., 2020]. These AIDs equate to SG cochlear place frequencies of 590 Hz, 646 Hz, and 713 Hz, respectively.

For Case 1 (460º/498 Hz), the most apical contact (E1) is within the region of functional acoustic hearing for the EAS simulation. Stimulation from contacts within the region of functional acoustic hearing may interfere with the neural response to the acoustic signal, known as electric-on-acoustic masking [Stronks et al, 2010; Lin et al., 2011; Stronks et al., 2012; Krüger, Büchner, & Nogueira, 2017; Imsiecke et al., 2020; Kipping, Krüger, & Nogueira, 2020]. Our place-based mapping procedure attempts to limit electric-on-acoustic masking by reducing the stimulation level below detection for contacts within the region of functional acoustic hearing. For the present study, this was simulated for Case 1 by omitting the vocoder band associated with E1 for the EAS simulations. That is, an 11-channel vocoder (E2-E12) was used for the EAS-Strict and EAS-LFshift conditions.

Strict versus Low-Frequency Shift Place-Based Mapping Procedures

For both the Strict-PB or LFshift-PB mapping procedures, the filter frequencies were adjusted to align the input with the cochlear place frequencies for contacts residing within the mid-frequency cochlear region (i.e., 1 – 3 kHz). The high-frequency information was distributed across the remaining channels for contacts in the basal cochlear region (i.e., > 3 kHz). Thus, filter frequencies for the Strict-PB and LFshift-PB maps were similar for contacts in the mid- to high-frequency cochlear regions.

The Strict-PB and LFshift-PB procedures differed in the distribution of low-frequency information. For the Strict-PB procedure, the filter frequencies were adjusted to align the input with the cochlear place frequencies for contacts apical to the 1 kHz cochlear region. Low-frequency information that was outside of the filter boundary for E1 was discarded. For the LFshift-PB procedure, the filters for the contacts apical to the 1 kHz cochlear region were widened to provide more low-frequency information. Thus, Strict-PB maps aligned with cochlear tonotopicity yet discarded lower frequency information; LFshift-PB maps provided more of the speech spectrum yet spectrally shifted the low-frequency information. Figure 1 shows the spectral shifts in low-frequency information with the LFshift-PB map as compared to the Strict-PB map for each case.

Procedure

Participants completed a task of masked sentence recognition while listening to either a CI or EAS simulation for one of the three cases. Listeners were seated in a quiet room. Stimuli were routed through an external sound card (M-AUDIO, M-Track 2×2) and presented diotically over headphones (Sennheiser, HD 280 Pro). The experiment was controlled by a custom MATLAB script.

Speech recognition was evaluated using an adaptive, ascending signal-to-noise ratio (SNR) procedure, as previously described by Buss et al. [2015] and used in our other vocoder experiments by Dillon et al. [2021, 2022]. Briefly, the AzBio sentences [Spahr et al., 2012] were presented in a 10-talker masker. The masker level was 60 dB SPL, and the starting level for the target was 0 dB SNR. The listener was asked to repeat the target and was scored for each word correctly repeated. Target level increased in 5 dB steps until the listener correctly repeated all the words in the sentence or the maximum SNR was reached (i.e., 20 dB). The ascending procedure was completed for each of the 20 sentences within the list. Feedback was not provided.

Twenty listeners provided data for each case (3 cases, 60 total listeners). For each case, four conditions were evaluated (i.e., CI-Strict, CI-LFshift, EAS-Strict, and EAS-LFshift). Listeners were randomized to listen with either a CI or EAS simulation. For each device simulation, half of the participants listened with the Strict-PB map first, and half listened to the LFshift-PB map first to control for potential learning effects.

Data analysis

The proportion of correctly repeated words at each SNR was fitted with a three-parameter logit function (i.e., mean, slope, and asymptote) to generate the psychometric functions for the plots. For the data analysis, proportion correct values were restricted within 0.001–0.999, and a logit transformation was applied to normalize the variance [Oleson, Brown, & McCreery, 2019]. The primary aim was to determine whether presenting low-frequency information with spectral shifts (LFshift-PB maps) offer an advantage over discarding low-frequency information (Strict-PB maps) when the mid-frequency information is aligned with cochlear place for electrode arrays at shallow AIDs. A linear mixed model evaluated the main effects of sex, case (Cases 1, 2, and 3), condition (CI-LFshift, CI-Strict, EAS-LFshift, and EAS-Strict), and SNR (0, 5, 10, 15, and 20 dB), and the 2-way and 3-way interactions of case, condition, and SNR using the lme function in R statistical software (R Core Team, 2021), with a random intercept for each listener. Case and condition were entered as factors. The case with the shallowest AID, Case 3 (335º/987 Hz), was the reference case due to the prediction that the performance benefit for Strict-PB over LFshift-PB maps would reverse for the case with the shallowest AID. SNR was mean centered on 10 dB.

The second aim was to evaluate whether the addition of acoustic low-frequency cues with EAS influence performance differences with the Strict-PB versus LFshift-PB maps as a function of AID. Reduced models assessed the patterns of performance for the EAS and CI simulations individually to evaluate performance with Strict-PB versus LFshift-PB with and without acoustic low-frequency cues. The reduced models included main effects of case, map (Strict-PB and LFshift-PB), and SNR, as well as the associated interactions. Significance was defined as ∝ < 0.05.

Results

Figure 2 shows psychometric functions fitted to the mean proportion correct. Data for the three cases are shown in separate panels. Mean proportion correct at each SNR is indicated with circles for the CI simulations and diamonds for the EAS simulations. Filled symbols and solid lines indicate performance and fit with the LFshift-PB maps; open symbols and dashed lines indicate performance and fit with the Strict-PB maps. Functions fitted to data of individual listeners are plotted in Supplemental Figure 1.

Figure 2:

Figure 2:

Open in a new tab

Psychometric functions fitted to mean proportion correct data for each case. Symbols show mean proportion correct at each SNR, and lines indicate fits for condition (i.e., CI-LFshift, CI-Strict, EAS-LFshift, EAS-Strict), as specified in the legend. The angular insertion depth and cochlear place frequency of the most apical electrode contact is provided for each case.

Performance between place-based mapping procedures across cases

The first model evaluated whether the LFshift-PB maps offered an advantage for masked speech recognition over the Strict-PB maps across simulations of electrode arrays at shallow AIDs. Table 1 lists the coefficients from the full model. There was a significant main effect of condition (F(3,519) = 5.87, p = 0.001). As compared the CI-LFshift condition, performance was significantly better with the EAS-LFshift and EAS-Strict conditions (p < 0.001), likely due to the addition of the acoustic low-frequency cues (see below). The differences in performance between conditions were more pronounced at the higher SNRs (i.e., ≥ 10 dB), with a significant interaction between condition and SNR (F(3,519) = 6.40, p < 0.001). Also, there was a significant interaction between case and condition (F(6,519) = 4.01, p = 0.001), indicating that the patterns of performance across the conditions differed for the three cases. Scores for the EAS-Strict and CI-LFshift conditions were relatively consistent across cases. For example, at 10 dB SNR the proportion correct as AID decreased (Case 1 to Case 3) was 0.71, 0.73, and 0.74 for EAS-Strict and 0.53, 0.54, and 0.49 for CI-LFshift. For the EAS-LFshift condition, performance improved as the AID decreased (i.e., 0.54, 0.60, and 0.70, respectively at 10 dB SNR). Conversely, performance declined as the AID decreased for the CI-Strict condition (i.e., 0.61, 0.56, and 0.50, respectively at 10 dB SNR). There was a significant 3-way interaction (F(6,519) = 3.99, p = 0.001) between case, condition, and SNR, indicating that the case-by-condition interaction was most pronounced at high SNRs. There was no significant main effect of sex (F(1,56) = 0.78, p = 0.381); this variable was removed from subsequent models.

Table 1:

Regression coefficients from the full LMM that evaluated the main effects of sex, case (Cases 1, 2, and 3), condition (CI-LFshift, CI-Strict, EAS-LFshift, and EAS-Strict), and SNR (0, 5, 10, 15, and 20 dB), and the 2-way and 3-way interactions of case, condition, and SNR. Significant results are indicated in bold and italics. Case 3 (335º/987 Hz) and CI-LFshift served as reference conditions. SNR was mean centered on 10 dB.

Coefficient SE DF t-value p-value
Sex −0.18 .20 56 −0.88 0.381
Case (Case 1) 0.22 0.33 56 0.66 0.514
Case (Case 2) 0.19 0.34 56 0.57 0.570
Condition (CI-Strict) −0.03 0.19 519 −0.15 0.877
Condition (EAS-LFshift) 1.26 0.33 519 3.83 <0.001
Condition (EAS-Strict) 1.23 0.33 519 3.74 <0.001
SNR 0.27 0.02 519 14.46 <0.001
Case 1: CI-Strict 0.64 0.27 519 2.40 0.017
Case 2: CI-Strict 0.30 0.27 519 1.10 0.271
Case 1: EAS-LFshift −1.08 0.46 519 −2.33 0.020
Case 2: EAS-LFshift −0.74 0.46 519 −1.58 0.114
Case 1: EAS-Strict −0.21 0.46 519 −0.44 0.658
Case 2: EAS-Strict 0.06 0.46 519 0.12 0.902
Case 1: SNR −0.05 0.03 519 −2.01 0.045
Case 2: SNR −0.02 0.03 519 −0.78 0.435
CI-Strict: SNR 0.00 0.03 519 0.14 0.886
EAS-LFshift: SNR 0.09 0.03 519 3.45 <0.001
EAS-Strict: SNR 0.08 0.03 519 2.82 0.005
Case 1: CI-Strict: SNR 0.04 0.04 519 1.03 0.304
Case 2: CI-Strict: SNR 0.04 0.04 519 1.09 0.276
Case 1: EAS-LFshift: SNR −0.09 0.04 519 −2.38 0.018
Case 2: EAS-LFshift: SNR −0.10 0.04 519 −2.65 0.008
Case 1: EAS-Strict: SNR 0.04 0.04 519 1.02 0.306
Case 2: EAS-Strict: SNR 0.05 0.04 519 1.35 0.177
Open in a new tab

Patterns of performance by device across cases

To further evaluate the interaction between case and condition, reduced models evaluated the EAS and CI simulation data separately to assess performance between the Strict-PB and LFshift-PB maps with and without acoustic low-frequency cues.

For the CI simulations, it was predicted that performance with the strict map (CI-Strict) would decline with decreases in AID and that listeners would experience better performance with the LFshift-PB map (CI-LFshift). Table 2 lists the coefficients for the model that assessed the data from the CI-Strict and CI-LFshift conditions. The main effects of case and map were non-significant (p ≥ 0.172). There was a significant interaction between case and map (F(2,261) = 6.19, p = 0.002), indicating differences in the patterns of performance with the Strict-PB versus LFshift-PB maps across the cases. Review of the fixed effects demonstrate that listeners experienced significantly better performance with the Strict-PB map for Cases 1 and 2 (p ≤ 0.004). Taken together, these data indicate that our Strict-PB procedure may not be detrimental for the performance for CI users with electrode arrays at AIDs of ≥ 335º. Additionally, there was a significant 3-way interaction between case, map, and SNR (F(2,261) = 8.82, p < 0.001), indicating that the interaction between effects of map and case was largest at high SNRs.

Table 2:

Regression coefficients from the reduced LMM for the CI simulations that evaluated the main effects of case (Cases 1, 2, and 3), map (Strict-PB and LFshift-PB), and SNR (0, 5, 10, 15, and 20 dB), and their 2-way and 3-way interactions. Significant results are indicated in bold and italics. Case 3 (335º/987 Hz) and the LFshift-PB map served as reference conditions. SNR was mean centered on 10 dB.

Coefficient SE DF t-value p-value
Case (Case 1) 0.86 0.54 27 1.58 0.125
Case (Case 2) 0.95 0.54 27 1.76 0.089
Map 0.03 0.19 261 0.15 0.880
SNR 0.33 0.04 261 7.64 <0.001
Case 1: Map −0.88 0.28 261 −3.19 0.002
Case 2: Map −0.79 0.28 261 −2.88 0.004
Case 1: SNR 0.11 0.06 261 1.85 0.065
Case 2: SNR 0.18 0.06 261 2.95 0.003
Map: SNR 0.02 0.03 261 0.62 0.537
Case 1: Map: SNR −0.13 0.04 261 −3.31 0.001
Case 2: Map: SNR −0.15 0.04 261 −3.89 <0.001
Open in a new tab

For the EAS simulations, we predicted better performance with Strict-PB maps (EAS-Strict) than LFshift-PB maps (EAS-LFshift) over a wider range of AIDs due to the acoustic low-frequency information serving as an anchor to cochlear tonotopicity. Table 3 lists the coefficients for the model that assessed the data from the EAS-Strict and EAS-LFshift conditions. There was a significant main effect of case (F(2,27) = 5.06, p = 0.014), with fixed effects demonstrating a significant difference in performance between Cases 1 and 3 (p = 0.004), but not for Cases 2 and 3 (p = 0.131). There was a significant interaction between case and map (F(2,261) = 3.07, p = 0.048), with fixed effects demonstrating different patterns of results for the two maps for Case 1 as compared to Case 3 (p = 0.014) but not for Case 2 as compared to Case 3 (p = 0.258). These data reflect the fact that listeners experienced an improvement with the LFshift-PB maps as AID decreased, which was similar to the performance observed with the Strict-PB maps for the shallower cases (i.e., Cases 2 and 3). These results suggest that EAS users with functional acoustic hearing at .125 and .25 kHz and an electrode array placed within 389 – 335º may experience similar benefit with Strict-PB versus LFshift-PB maps. Better performance with a Strict-PB map may be experienced for electrode arrays at deeper AIDs (e.g., 460º).

Table 3:

Regression coefficients from the LMM for the EAS simulations that evaluated the main effects of case (Cases 1, 2, and 3), map (Strict-PB and LFshift-PB), and SNR (0, 5, 10, 15, and 20 dB), and their 2-way and 3-way interactions. Significant results are indicated in bold and italics. Case 3 (335º/987 Hz) and the LFshift-PB map served as reference conditions. SNR was mean centered on 10 dB.

Coefficient SE DF t-value p-value
Case (Case 1) 1.45 0.46 27 3.18 0.004
Case (Case 2) 0.71 0.46 27 1.56 0.131
Map 0.03 0.18 261 0.16 0.874
SNR 0.28 0.04 261 6.85 <0.001
Case 1: Map −0.64 0.26 261 −2.48 0.014
Case 2: Map −0.30 0.26 261 −1.13 0.258
Case 1: SNR 0.02 0.06 261 0.41 0.680
Case 2: SNR 0.06 0.06 261 1.06 0.289
Map: SNR −0.00 0.03 261 −0.15 0.883
Case 1: Map: SNR −0.04 0.04 261 −1.06 0.290
Case 2: Map: SNR −0.04 0.04 261 −1.12 0.262
Open in a new tab

Discussion

There is a growing interest in individualizing the mapping of electric stimulation for CI and EAS users, such as with place-based mapping procedures; however, the optimal procedure by which audiologists perform place-based mapping in clinical settings remains unclear. The Strict-PB procedure described herein aligns the electric filter frequencies to the cochlear place frequencies up to at least 3 kHz and distributes the remaining high frequency information across the basal contacts. The effectiveness of strictly aligning the electric filters to cochlear tonotopicity may be limited for recipients of electrode arrays at shallow AIDs due to the loss of low-frequency information [Fu & Shannon, 1999b; Faulkner et al., 2003; Başkent & Shannon, 2005]. The LFshift-PB procedure aligns the critical speech information (e.g., 1–3 kHz) and compresses lower and higher frequency information to provide more of the speech spectrum, essentially sacrificing strict tonotopicity for full spectral representation. The present study sought to inform procedural strategies for place-based mapping by comparing acute masked speech recognition for CI and EAS simulations with Strict-PB versus LFshift-PB maps for three example cases with a range of AIDs. Generally, listeners experienced better performance with Strict-PB maps than LFshift-PB maps for Case 1 (460º/498 Hz) and Case 2 (389º/728 Hz), which had the largest frequency-to-place mismatches on the apical channels with the LFshift-PB maps. For Case 3 (335º/987 Hz), listeners of CI and EAS simulations experienced similar performance with Strict-PB maps and LFshift-PB maps. This is not entirely surprising since the two maps only differed in the frequency information provided by E1. These data provide preliminary support for the use of the Strict-PB procedure over the LFshift-PB procedure for CI and EAS users – at least for recipients of electrode arrays at AIDs ≥ 335º.

The effects of listening with a Strict-PB map or a LFshift-PB map differed for the CI and EAS simulations, indicating different performance benefits with Strict-PB versus LFshift-PB maps with the addition of acoustic low-frequency cues. For the CI simulations, performance with the Strict-PB maps declined with decreases in AID, while the performance with the LFshift-PB maps was relatively consistent across the three cases. The observation of degraded performance with the Strict-PB maps as AID decreased corroborates previous data demonstrating declines in speech recognition in quiet at shallow insertion depths due to the loss of low-frequency information [Fu & Shannon, 1999b; Faulkner et al., 2003; Başkent & Shannon, 2005]. Interestingly, performance in the present study was relatively consistent across cases with the LFshift-PB maps – despite differences in the number of low-frequency channels with spectral shifts. Case 1 had more channels with spectral shifts (n=3) than Case 2 (n=2) or Case 3 (n=1). In other words, there was no performance benefit associated with the inclusion of low-frequency information when the mid-frequency information is aligned with cochlear place.

In contrast to CI simulations, results from the EAS simulations indicate that performance with the Strict-PB maps was relatively consistent across the three cases despite differences in the available electric frequency information. Whereas performance in the CI-Strict condition declined with decreases in AID, consistent performance in the EAS-Strict condition across cases is likely due to the inclusion of low-frequency acoustic cues – though limited to 125 and 250 Hz. These data corroborate previous EAS simulation and user data that demonstrate superior performance with strict place-based maps over spectrally shifted maps – even when there is a frequency information gap between the acoustic and electric outputs [Fu et al., 2017; Willis et al., 2020; Dillon et al., 2021, 2023]. The present data extend that previous finding by showing that closing the frequency information gap between the acoustic and electric outputs by spectrally shifting some low-frequency information with a LFshift-PB map does not provide superior performance over Strict-PB maps that create a frequency information gap at least with acute exposure. Taken together, these data suggest that incorporating the individual differences in the placement of the electrode array relative to cochlear tonotopicity into the mapping of EAS devices may support better performance than with current default procedures, although is it unclear whether extended listening experience ameliorates detrimental effects of electric frequency-to-place mismatches.

Another finding from the EAS simulations was the emerging benefit with the LFshift-PB maps as AID decreased. This performance benefit may have been due to decreases in the number of channels with spectrally shifted low-frequency information. However, this possibility seems unlikely since Cases 1 and 2 had two channels with spectral shifts and Case 3 had one channel. As a reminder, the EAS simulations for Case 1 used an 11-channel vocoder to simulate the lowered stimulation of E1 to attempt to limit electric-on-acoustic masking. It seems more likely that the increasing benefit of the LFshift-PB map with decreases in AID was due to the inclusion of some spectrally shifted low-frequency information that was not redundant with the acoustic information. The present data cannot differentiate effects of spectrally shifting low-frequency electric information versus the redundancy of acoustic and electric information on the performance with the EAS-LFshift condition for Cases 1 and 2. Notably, performance with the LFshift-PB map did not exceed that with the Strict-PB map for the case with the shallowest insertion depth. This suggests that Strict-PB maps may support better performance for EAS users than LFshift-PB maps – at least using the AIDs, settings, and acoustic hearing simulated in the present study.

An important consideration of the present data is that acute performance was evaluated for listeners with normal hearing who did not have previous experience with vocoded speech or auditory training with the simulations. CI and EAS users may acclimate to spectral shifts with listening experience and/or auditory training [Rosen, Faulkner, & Wilkinson, 1999; Fu, Shannon, & Galvin, 2002; Fu & Galvin, 2003; Svirsky et al, 2004; Fu et al., 2005; Faulkner, 2006; Li & Fu, 2007; Reiss et al., 2007; Li, Galvin, & Fu, 2009; Sagi et al., 2010; Smalt et al., 2013; Reiss et al., 2014; Svirsky et al., 2015; Vermeire et al., 2015; James et al., 2019]. For example, Li and colleagues [2009] observed that participants with normal hearing listening to CI simulations acclimated to spectral shifts of ≤ 6 mm with 5 days of listening experience. More recently, James and colleagues [2019] reported a steep increase in speech recognition between 1-day and 1-month post-activation for CI users with spectrally shifted maps. As such, the present results may not be representative of performance that would have been obtained if participants had completed auditory training or had longer listening experience with the simulations, and the acute performance differences observed in the present study may not represent the long-term performance of actual CI and EAS users. On the other hand, acclimatization to spectrally shifted maps does not always allow for users to overcome performance deficits [Reiss et al., 2007; Sagi et al., 2010; Reiss et al., 2014; Svirsky et al., 2015], and there are individual differences in the ability to acclimate to spectrally shifted maps [Smith & Winn, 2021]. Preliminary data from CI users with normal hearing in the contralateral ear also suggest that while speech recognition may improve with spectrally shifted maps, consistent with acclimatization to frequency shifts, the majority of recipients may continue to perceive the frequency-to-place mismatches after 5 years of listening experience [Karoui et al., 2019].

There are other limitations of the present study worth consideration. Performance differences between Strict-PB and LFshift-PB maps may be reduced in CI and EAS users due to current spread and channel interactions [Fu & Shannon, 1999a; Friesen et al., 2001]. Also, we cannot rule out the potential influence of high-frequency spectral shifts on the observed patterns of performance. There was an increase in the number of channels with spectrally shifted high-frequency information as the AID decreased. These shifts may have contributed to the decline in performance in the CI-Strict condition as AID decreased. Investigation is needed to determine performance differences between the present place-based mapping procedures and procedures that strictly align all frequency information to cochlear tonotopicity, such as deactivating contacts that exceed the current upper filter frequency limits [e.g., 8.5 kHz; see Jiam et al., 2019].

Conclusions

The present CI and EAS simulation data provide preliminary support for the use of a Strict-PB procedure for CI and EAS users with AIDs ≥ 335º. Ongoing work is evaluating performance differences over time for CI and EAS users with Strict-PB maps as compared to default maps with different magnitudes of spectral shifts.

Supplementary Material

Supplemental Table 1
Supplemental Figure 1

Acknowledgements:

Stacey Kane, Meredith Rooth, Margaret Richter, Samantha Scharf, Kathryn Young, and Shannon Culbertson assisted with subject recruitment and/or data collection.

Funding Sources:

This work was funded in part by a clinical research grant provided by the Department of Otolaryngology/Head & Neck Surgery at the University of North Carolina at Chapel Hill. Funding was also provided by the National Institute on Deafness and Other Communication Disorders of the National Institutes of Health under Award Number R21DC018389. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The data collection and preparation of the manuscript was conducted by the authors.

Footnotes

Dissertation: A preprint version of this manuscript (Dillon, 2022) is available at the Carolina Digital Repository: https://cdr.lib.unc.edu/concern/dissertations/2f75rj76t

Dillon, M.T. (2022). Place-based mapping with electric-acoustic stimulation. [Doctoral dissertation, University of North Carolina at Chapel Hill]. ProQuest Dissertation Publication.

Statement of Ethics: The University of North Carolina at Chapel Hill (UNC) Biomedical Research Institutional Review Board reviewed and approved the study protocol (UNC IRB approval number 86–0059). Participants provided written informed consent prior to completing the study procedures.

Conflicts of Interest Statement: Margaret Dillon and Emily Buss are supported by a research grant provided to the university by MED-EL Corporation. Brendan O’Connell is a consultant for Advanced Bionics Corporation, MED-EL Corporation, and Johnson and Johnson.

Data Availability Statement:

Data will be made available by emailing the corresponding author (Margaret Dillon, mdillon@med.unc.edu). There are no legal or ethical reasons that prohibit the sharing of the data.

References

  1. ANSI/ASA S3.5–1997. (1997). ANSI/ASA S3.5–1997 (R2017) - Methods for Calculation of the Speech Intelligibility Index Retrieved December 20, 2021, from
  2. Başkent D, & Shannon RV (2003). Speech recognition under conditions of frequency-place compression and expansion. The Journal of the Acoustical Society of America, 113(4 Pt 1), 2064–2076. 10.1121/1.1558357 [DOI] [PubMed] [Google Scholar]
  3. Başkent D, & Shannon RV (2005). Interactions between cochlear implant electrode insertion depth and frequency-place mapping. The Journal of the Acoustical Society of America, 117(3 Pt 1), 1405–1416. 10.1121/1.1856273 [DOI] [PubMed] [Google Scholar]
  4. Başkent D, & Shannon RV (2007). Combined effects of frequency compression-expansion and shift on speech recognition. Ear and Hearing, 28(3), 277–289. 10.1097/AUD.0b013e318050d398 [DOI] [PubMed] [Google Scholar]
  5. Buss E, Calandruccio L, & Hall JW (2015). Masked sentence recognition assessed at ascending target-to-masker ratios: modest effects of repeating stimuli. Ear and Hearing, 36(2), e14–22. 10.1097/AUD.0000000000000113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Canfarotta MW, Dillon MT, Buss E, Pillsbury HC, Brown KD, & O’Connell BP (2020). Frequency-to-Place Mismatch: Characterizing Variability and the Influence on Speech Perception Outcomes in Cochlear Implant Recipients. Ear and Hearing, 41(5), 1349–1361. 10.1097/AUD.0000000000000864 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dillon MT, Canfarotta MW, Buss E, Hopfinger J, & O’Connell BP (2021). Effectiveness of Place-based Mapping in Electric-Acoustic Stimulation Devices. Otology & Neurotology, 42(1), 197–202. 10.1097/MAO.0000000000002965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dillon MT, Canfarotta MW, Buss E, Rooth MA, Richter ME, Overton AB, … O’Connell BP (2023). Influence of Electric Frequency-to-Place Mismatches on the Early Speech Recognition Outcomes for Electric-Acoustic Stimulation Users. American Journal of Audiology, 1–10. 10.1044/2022_AJA-21-00254 [DOI] [PMC free article] [PubMed]
  9. Dillon MT, O’Connell BP, Canfarotta MW, Buss E, & Hopfinger J (2022). Effect of Place-Based Versus Default Mapping Procedures on Masked Speech Recognition: Simulations of Cochlear Implant Alone and Electric-Acoustic Stimulation. American Journal of Audiology, 1–16. 10.1044/2022_AJA-21-00123 [DOI] [PMC free article] [PubMed]
  10. Dillon MT (2022). Place-based mapping with electric-acoustic stimulation [Doctoral dissertation, University of North Carolina at Chapel Hill]. ProQuest Dissertation Publication. 10.17615/g6m4-5d96 [DOI] [Google Scholar]
  11. Dorman MF, Loizou PC, & Rainey D (1997). Simulating the effect of cochlear-implant electrode insertion depth on speech understanding. The Journal of the Acoustical Society of America, 102(5 Pt 1), 2993–2996. 10.1121/1.420354 [DOI] [PubMed] [Google Scholar]
  12. Faulkner A (2006). Adaptation to distorted frequency-to-place maps: implications of simulations in normal listeners for cochlear implants and electroacoustic stimulation. Audiology & Neuro-Otology, 11 Suppl 1, 21–26. 10.1159/000095610 [DOI] [PubMed] [Google Scholar]
  13. Faulkner A, Rosen S, & Stanton D (2003). Simulations of tonotopically mapped speech processors for cochlear implant electrodes varying in insertion depth. The Journal of the Acoustical Society of America, 113(2), 1073–1080. 10.1121/1.1536928 [DOI] [PubMed] [Google Scholar]
  14. Friesen LM, Shannon RV, Baskent D, & Wang X (2001). Speech recognition in noise as a function of the number of spectral channels: comparison of acoustic hearing and cochlear implants. The Journal of the Acoustical Society of America, 110(2), 1150–1163. 10.1121/1.1381538 [DOI] [PubMed] [Google Scholar]
  15. Fu Q-J, & Shannon RV (1999a). Effects of electrode location and spacing on phoneme recognition with the Nucleus-22 cochlear implant. Ear and Hearing, 20(4), 321–331. 10.1097/00003446-199908000-00005 [DOI] [PubMed] [Google Scholar]
  16. Fu Q-J, & Shannon RV (1999b). Recognition of spectrally degraded and frequency-shifted vowels in acoustic and electric hearing. The Journal of the Acoustical Society of America, 105(3), 1889–1900. 10.1121/1.426725 [DOI] [PubMed] [Google Scholar]
  17. Fu Q-J, Chinchilla S, Nogaki G, & Galvin JJ (2005). Voice gender identification by cochlear implant users: the role of spectral and temporal resolution. The Journal of the Acoustical Society of America, 118(3 Pt 1), 1711–1718. 10.1121/1.1985024 [DOI] [PubMed] [Google Scholar]
  18. Fu Q-J, & Galvin JJ (2003). The effects of short-term training for spectrally mismatched noise-band speech. The Journal of the Acoustical Society of America, 113(2), 1065–1072. 10.1121/1.1537708 [DOI] [PubMed] [Google Scholar]
  19. Fu Q-J, Galvin JJ, & Wang X (2017). Integration of acoustic and electric hearing is better in the same ear than across ears. Scientific Reports, 7(1), 12500. 10.1038/s41598-017-12298-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Fu Q-J, Shannon RV, & Galvin JJ (2002). Perceptual learning following changes in the frequency-to-electrode assignment with the Nucleus-22 cochlear implant. The Journal of the Acoustical Society of America, 112(4), 1664–1674. 10.1121/1.1502901 [DOI] [PubMed] [Google Scholar]
  21. Imsiecke M, Büchner A, Lenarz T, & Nogueira W (2020). Psychoacoustic and electrophysiological electric-acoustic interaction effects in cochlear implant users with ipsilateral residual hearing. Hearing Research, 386, 107873. 10.1016/j.heares.2019.107873 [DOI] [PubMed] [Google Scholar]
  22. James CJ, Karoui C, Laborde M-L, Lepage B, Molinier C-É, Tartayre M, … Fraysse B (2019). Early sentence recognition in adult cochlear implant users. Ear and Hearing, 40(4), 905–917. 10.1097/AUD.0000000000000670 [DOI] [PubMed] [Google Scholar]
  23. Jiam NT, Gilbert M, Cooke D, Jiradejvong P, Barrett K, Caldwell M, & Limb CJ (2019). Association Between Flat-Panel Computed Tomographic Imaging-Guided Place-Pitch Mapping and Speech and Pitch Perception in Cochlear Implant Users. JAMA Otolaryngology-- Head & Neck Surgery, 145(2), 109–116. 10.1001/jamaoto.2018.3096 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jin SH, Nelson PB. Interrupted speech perception: The effects of hearing sensitivity and frequency resolution. J Acoust Soc Am 2010;128(2):881–889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Karoui C, James C, Barone P, Bakhos D, Marx M, & Macherey O (2019). Searching for the Sound of a Cochlear Implant: Evaluation of Different Vocoder Parameters by Cochlear Implant Users With Single-Sided Deafness. Trends in Hearing, 23, 2331216519866029. 10.1177/2331216519866029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kipping D, Krüger B, & Nogueira W (2020). The role of electroneural versus electrophonic stimulation on psychoacoustic electric-acoustic masking in cochlear implant users with residual hearing. Hearing Research, 395, 108036. 10.1016/j.heares.2020.108036 [DOI] [PubMed] [Google Scholar]
  27. Krüger B, Büchner A, & Nogueira W (2017). Simultaneous masking between electric and acoustic stimulation in cochlear implant users with residual low-frequency hearing. Hearing Research, 353, 185–196. 10.1016/j.heares.2017.06.014 [DOI] [PubMed] [Google Scholar]
  28. Landsberger DM, Svrakic M, Roland JT, & Svirsky M (2015). The relationship between insertion angles, default frequency allocations, and spiral ganglion place pitch in cochlear implants. Ear and Hearing, 36(5), e207–13. 10.1097/AUD.0000000000000163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lenarz T, Buechner A, Lesinski-Schiedat A, Timm M, & Salcher R (2020). Hearing preservation with a new atraumatic lateral wall electrode. Otology & Neurotology, 41(8), e993–e1003. 10.1097/MAO.0000000000002714 [DOI] [PubMed] [Google Scholar]
  30. Li T, & Fu Q-J (2007). Perceptual adaptation to spectrally shifted vowels: training with nonlexical labels. Journal of the Association for Research in Otolaryngology, 8(1), 32–41. 10.1007/s10162-006-0059-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Li T, & Fu Q-J (2010). Effects of spectral shifting on speech perception in noise. Hearing Research, 270(1–2), 81–88. 10.1016/j.heares.2010.09.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Li T, Galvin JJ, & Fu Q-J (2009). Interactions between unsupervised learning and the degree of spectral mismatch on short-term perceptual adaptation to spectrally shifted speech. Ear and Hearing, 30(2), 238–249. 10.1097/AUD.0b013e31819769ac [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lin P, Turner CW, Gantz BJ, Djalilian HR, & Zeng F-G (2011). Ipsilateral masking between acoustic and electric stimulations. The Journal of the Acoustical Society of America, 130(2), 858–865. 10.1121/1.3605294 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. O’Connell BP, Hunter JB, Gifford RH, Rivas A, Haynes DS, Noble JH, & Wanna GB (2016). Electrode location and audiologic performance after cochlear implantation: A comparative study between nucleus CI422 and CI512 electrode arrays. Otology & Neurotology, 37(8), 1032–1035. 10.1097/MAO.0000000000001140 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Oleson JJ, Brown GD, & McCreery R (2019). The evolution of statistical methods in speech, language, and hearing sciences. Journal of Speech, Language, and Hearing Research, 62(3), 498–506. 10.1044/2018_JSLHR-H-ASTM-18-0378 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Qin MK, Oxenham AJ. Effect of simulated cochlear-implant proces sing on speech reception in fluctuating maskers. J Acoust Soc Am 2003;114(1):446–454. [DOI] [PubMed] [Google Scholar]
  37. Reiss LAJ, Turner CW, Karsten SA, & Gantz BJ (2014). Plasticity in human pitch perception induced by tonotopically mismatched electro-acoustic stimulation. Neuroscience, 256, 43–52. 10.1016/j.neuroscience.2013.10.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Reiss LAJ, Turner CW, Erenberg SR, & Gantz BJ (2007). Changes in pitch with a cochlear implant over time. Journal of the Association for Research in Otolaryngology, 8(2), 241–257. 10.1007/s10162-007-0077-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Rosen S, Faulkner A, & Wilkinson L (1999). Adaptation by normal listeners to upward spectral shifts of speech: implications for cochlear implants. The Journal of the Acoustical Society of America, 106(6), 3629–3636. [DOI] [PubMed] [Google Scholar]
  40. Sagi E, Fu Q-J, Galvin JJ, & Svirsky MA (2010). A model of incomplete adaptation to a severely shifted frequency-to-electrode mapping by cochlear implant users. Journal of the Association for Research in Otolaryngology, 11(1), 69–78. 10.1007/s10162-009-0187-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shannon RV, Zeng FG, & Wygonski J (1998). Speech recognition with altered spectral distribution of envelope cues. The Journal of the Acoustical Society of America, 104(4), 2467–2476. [DOI] [PubMed] [Google Scholar]
  42. Smalt CJ, Gonzalez-Castillo J, Talavage TM, Pisoni DB, & Svirsky MA (2013). Neural correlates of adaptation in freely-moving normal hearing subjects under cochlear implant acoustic simulations. Neuroimage, 82, 500–509. 10.1016/j.neuroimage.2013.06.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Smith ML, & Winn MB (2021). Individual variability in recalibrating to spectrally shifted speech: implications for cochlear implants. Ear and Hearing 10.1097/AUD.0000000000001043 [DOI] [PMC free article] [PubMed]
  44. Spahr AJ, Dorman MF, Litvak LM, Van Wie S, Gifford RH, Loizou PC, … Cook S (2012). Development and validation of the AzBio sentence lists. Ear and Hearing, 33(1), 112–117. 10.1097/AUD.0b013e31822c2549 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Stakhovskaya O, Sridhar D, Bonham BH, & Leake PA (2007). Frequency map for the human cochlear spiral ganglion: implications for cochlear implants. Journal of the Association for Research in Otolaryngology, 8(2), 220–233. 10.1007/s10162-007-0076-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Stronks HC, Prijs VF, Chimona TS, Grolman W, & Klis SFL (2012). Spatial overlap of combined electroacoustic stimulation determines the electrically evoked response in the guinea pig cochlea. Otology & Neurotology, 33(9), 1535–1542. 10.1097/MAO.0b013e318271c0b6 [DOI] [PubMed] [Google Scholar]
  47. Stronks HC, Versnel H, Prijs VF, & Klis SFL (2010). Suppression of the acoustically evoked auditory-nerve response by electrical stimulation in the cochlea of the guinea pig. Hearing Research, 259(1–2), 64–74. 10.1016/j.heares.2009.10.004 [DOI] [PubMed] [Google Scholar]
  48. Svirsky MA, Silveira A, Neuburger H, Teoh S-W, & Suárez H (2004). Long-term auditory adaptation to a modified peripheral frequency map. Acta Oto-Laryngologica, 124(4), 381–386. [PubMed] [Google Scholar]
  49. Svirsky Mario A, Talavage TM, Sinha S, Neuburger H, & Azadpour M (2015). Gradual adaptation to auditory frequency mismatch. Hearing Research, 322, 163–170. 10.1016/j.heares.2014.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Vermeire K, Landsberger DM, Van de Heyning PH, Voormolen M, Kleine Punte A, Schatzer R, & Zierhofer C (2015). Frequency-place map for electrical stimulation in cochlear implants: Change over time. Hearing Research, 326, 8–14. 10.1016/j.heares.2015.03.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Warren RM, Riener KR, Bashford JA, & Brubaker BS (1995). Spectral redundancy: intelligibility of sentences heard through narrow spectral slits. Perception & Psychophysics, 57(2), 175–182. 10.3758/bf03206503 [DOI] [PubMed] [Google Scholar]
  52. Willis S, Moore BCJ, Galvin JJ, & Fu Q-J (2020). Effects of noise on integration of acoustic and electric hearing within and across ears. Plos One, 15(10), e0240752. 10.1371/journal.pone.0240752 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Table 1
NIHMS1970713-supplement-Supplemental_Table_1.docx (20.1KB, docx)
Supplemental Figure 1
NIHMS1970713-supplement-Supplemental_Figure_1.docx (507.6KB, docx)

Data Availability Statement

Data will be made available by emailing the corresponding author (Margaret Dillon, mdillon@med.unc.edu). There are no legal or ethical reasons that prohibit the sharing of the data.

RESOURCES