Derivation of frequency importance functions for the AzBio sentences

The relationships between cochlear nerve health and AzBio sentence scores in quiet and noise in postlingually deafened adult cochlear implant users

Objectives

This study investigated the relationships between the cochlear nerve (CN) health and sentence-level speech perception outcomes measured in quiet and noise in postlingually deafened adult cochlear implant (CI) users.

Design

Study participants included 24 postlingually deafened adult CI users with a Cochlear® Nucleus™ device. For each participant, only one ear was tested. Neural health of the CN was assessed at three or four electrode locations across the electrode array using two parameters derived from results of the electrically evoked compound action potential (eCAP). One parameter was the phase locking value (PLV) which estimated neural synchrony in the CN. The other parameter was the sensitivity of the eCAP amplitude growth function (AGF) slope to changes in the interphase gap (IPG) of biphasic electrical pulses (i.e., the IPGEslope). Speech perception was tested using AzBio sentences in both quiet and a ten-talker babble background noise with +5 dB and +10 dB signal-to-noise ratios (SNR). IPGEslope and PLV values were averaged across electrodes for each subject, both with and without weighting by the frequency importance function (FIF) of the AzBio sentences. Pearson and Spearman correlations were used to assess the pairwise relationships between the IPGEslope, the PLV, and age. Multiple linear regression models with AzBio score as the outcome and the PLV and the IPGEslope as predictors were used to evaluate the associations between the three variables while controlling for age.

Results

The correlation between the IPGEslope and the PLV was negligible and not statistically significant. The PLV, but not the IPGEslope, differed significantly across electrodes, where the apical electrodes had larger PLVs (better neural synchrony) than the basal electrodes. The IPGEslope, but not the PLV, was significantly correlated with participant's age, where smaller IPGEslope values (poorer CN health) were associated with more advanced age. The PLV, but not the IPGEslope, was significantly associated with AzBio scores in noise, where larger PLVs predicted better speech perception in noise. Neither the PLV nor the IPGEslope was significantly associated with AzBio score in quiet. The result patterns remained the same regardless of whether the mean values of the IPGEslope and the PLV were weighted by the AzBio FIF.

Conclusions

The IPGEslope and the PLV quantify different aspects of CN health. The positive association between the PLV and AzBio scores suggests that neural synchrony is important for speech perception in noise in adult CI users. The lack of association between age and the PLV indicates that reduced neural synchrony in the CN is unlikely the primary factor accounting for the greater deficits in understanding speech in noise observed in elderly, as compared to younger, CI users.

Individualized estimation of the Speech Intelligibility Index for short sentences: Test-retest reliability

The speech intelligibility index (SII) model was modified to allow individualized parameters. These parameters included the relative weights of speech cues in five octave-frequency bands ranging from 0.25 to 4 kHz, i.e., the band importance function, and the transfer function that allows the SII to generate predictions on speech-recognition scores. A Bayesian adaptive procedure, the quick-band-importance-function (qBIF) procedure, was utilized to enable efficient estimation of the SII parameters from individual listeners. In two experiments, the SII parameters were estimated for 30 normal-hearing adults using Institute of Electrical and Electronics Engineers (IEEE) sentences at speech levels of 55, 65, and 75 dB sound pressure level (in Experiment I) and for 15 hearing-impaired (HI) adult listeners using amplified IEEE or AzBio sentences (in Experiment II). In both experiments, even without prior training, the estimated model parameters showed satisfactory reliability between two runs of the qBIF procedure at least one week apart. For the HI listeners, inter-listener variability in most estimated SII parameters was larger than intra-listener variability of the qBIF procedure.

The effect of speech material on the band importance function for Mandarin Chinese

Speech material influences the relative contributions of different frequency regions to intelligibility for English. In the current study, whether a similar effect of speech material is present for Mandarin Chinese was investigated. Speech recognition was measured using three speech materials in Mandarin, including disyllabic words, nonsense sentences, and meaningful sentences. These materials differed from one another in terms of the amount of contextual information and word frequency. The band importance function (BIF), as defined under the Speech Intelligibility Index (SII) framework, was used to quantify the contributions across frequency regions. The BIFs for the three speech materials were estimated from 16 adults who were native speakers of Mandarin. A Bayesian adaptive procedure was used to efficiently estimate the octave-frequency BIFs for the three materials for each listener. As the amount of contextual information increased, low-frequency bands (e.g., 250 and 500 Hz) became more important for speech recognition, consistent with English. The BIF was flatter for Mandarin than for comparable English speech materials. Introducing the language- and material-specific BIFs to the SII model led to improved predictions of Mandarin speech-recognition performance. Results suggested the necessity of developing material-specific BIFs for Mandarin.

Predicting Speech Recognition Using the Speech Intelligibility Index and Other Variables for Cochlear Implant Users

Purpose Although the speech intelligibility index (SII) has been widely applied in the field of audiology and other related areas, application of this metric to cochlear implants (CIs) has yet to be investigated. In this study, SIIs for CI users were calculated to investigate whether the SII could be an effective tool for predicting speech perception performance in a population with CI. Method Fifteen pre- and postlingually deafened adults with CI participated. Speech recognition scores were measured using the AzBio sentence lists. CI users also completed questionnaires and performed psychoacoustic (spectral and temporal resolution) and cognitive function (digit span) tests. Obtained SIIs were compared with predicted SIIs using a transfer function curve. Correlation and regression analyses were conducted on perceptual and demographic predictor variables to investigate the association between these factors and speech perception performance. Result Because of the considerably poor hearing and large individual variability in performance, the SII did not predict speech performance for this CI group using the traditional calculation. However, new SII models were developed incorporating predictive factors, which improved the accuracy of SII predictions in listeners with CI. Conclusion Conventional SII models are not appropriate for predicting speech perception scores for CI users. Demographic variables (aided audibility and duration of deafness) and perceptual-cognitive skills (gap detection and auditory digit span outcomes) are needed to improve the use of the SII for listeners with CI. Future studies are needed to improve our CI-corrected SII model by considering additional predictive factors. Supplemental Material https://doi.org/10.23641/asha.8057003.

Deactivating cochlear implant electrodes to improve speech perception: A computational approach

A potential bottleneck to improving speech perception performance in cochlear implant (CI) users is that some of their electrodes may poorly encode speech information. Several studies have examined the effect of deactivating poorly encoding electrodes on speech perception with mixed results. Many of these studies focused on identifying poorly encoding electrodes by some measure (e.g. electrode discrimination, pitch ordering, threshold, CT-guided, masked modulation detection), but provide inconsistent criteria about which electrodes, and how many, should be deactivated, and without considering how speech information becomes distributed across the electrode array. The present simulation study addresses this issue using computational approaches. Previously validated models were used to generate predictions of speech scores as a function of all possible combinations of active electrodes in a 22-electrode array in three groups of hypothetical subjects representative of relatively better, moderate, and poorer performing CI users. Using high-performance computing, over 500 million predictions were generated. Although deactivation of the poorest encoding electrodes sometimes resulted in predicted benefit, this benefit was significantly less relative to predictions resulting from model-optimized deactivations. This trend persisted when using novel stimuli (i.e. other than those used for optimization) and when using different processing strategies. Optimum electrode deactivation patterns produced an average predicted increase in word scores of 10% with some scores increasing by more than 20%. Optimum electrode deactivation patterns typically included 11 to 19 (out of 22) active electrodes, depending on the performance group. Optimal active electrode combinations were those that maximized discrimination of speech cues, maintaining 80%-100% of the physical span of the array. The present study demonstrates the potential for further improving CI users' speech scores with appropriate selection of active electrodes.