Psychoacoustic and Experimental Audiology - Completed Projects

Objective:

This project studies the effects of the upper-frequency boundary and of spectral warping on speech intelligibility among Cochlear Implant (CI) listeners, using a 12-channel implant, and normal hearing (NH) listeners.  This is important to determine how many basal channels are "free" for encoding spectral localization cues.

Results:

The results show that eight frequency channels and spectral content up to about 3 kHz are sufficient to transmit speech under unwarped conditions. If frequency warping was applied, the changes had to be limited ± 2 frequency channels to preserve good speech understanding. This outcome shows the range of allowed modifications for presenting spectral localization cues to CI listeners. About four channels were found to be "free" for encoding spectral localization cues

Application:

see the description of the CI-HRTF project

Funding:

FWF (Austrian Science Fund): Project #P18401-B15

Publications:

  • Goupell, M., Laback, B., Majdak, P., and Baumgartner, W. D. (2007). Effects of upper-frequency boundary and spectral warping on speech intelligibility in electrical stimulation, J. Acoust. Soc. Am. 123, 2295-2309.
  • Goupell, M. J., Laback, B., Majdak, P., and Baumgartner, W-D. (2007). Effect of frequency-place mapping on speech intelligibility: implications for a cochlear implant localization strategy, presented at Conference on Implantable Auditory Prostheses (CIAP), Lake Tahoe.
  • Goupell, M. J., Laback, B., Majdak, P., and Baumgartner, W-D. (2007). Effect of different frequency mappings on speech intelligibility for CI listeners, proceedings of DAGA 2007, Stuttgart.

Objective:

This study explores the adaptation of localization mechanisms to warping of spectral localization features, as required for CI listeners to map those features to their reduced electric stimulation range.

Methods and Results:

The effect of warping the stimulation range from 2.8 to 16 kHz to the range from 2.8 to 8.5 kHz was studied in normal-hearing listeners. Fifteen subjects participated in a long-time localization-training study, involving two-hour daily audio-visual training over a period of three weeks. The Test Group listened to frequency-warped stimuli, the Control Group to low-pass filtered stimuli (8.5 kHz). The Control Group showed an initial increase of localization error and essentially reached the baseline performance at the end of the training period. The Test Group showed a strong initial increase of localization error, followed by a steady improvement of performance, even though not reaching the baseline performance at the end of the training period. These results are promising with respect to the idea to present high-frequency spectral localization cues to the stimulation range available with CIs

Funding:

FWF (Austrian Science Fund): Project #P18401-B15

Publications:

  • Walder, T. (2010) Schallquellenlokalisation mittels Frequenzbereich-Kompression der Außenohrübertragungsfunktionen (englisch: Sound source localization with frequency-range compressed head-related transfer functions), Master thesis, Technical University of Graz & Kunstuniversität Graz.
  • Majdak, P., Walder, T., and Laback, B. (2011). Learning to Localize Band-Limited Sounds in Vertical Planes, presented at: 34st MidWinter Meeting of the Association for Research in Otolaryngology (ARO). Baltimore, Maryland.

Objective and Methods:

This study investigates the effect of the number of frequency channels on vertical place sound localization, especially front/back discrimination. This is important to determine how many of the basal-most channels/electrodes of a cochlear implant (CI) are needed to encode spectral localization cues. Normal hearing subjects listening to a CI simulation (the newly developed GET vocoder) will perform the experiment using the localization method developed in the subproject "Loca Methods". Learning effects will be studied by obtaining visual feedback.

Results:

Experiments are underway.

Application:

Knowing the number of channels required to encode spectral cues for localization in the vertical planes is an important step in the development of a 3-D localization strategy for CIs. 

Funding:

FWF (Austrian Science Fund): Project #P18401-B15

Publications:

  • Goupell, M., Majdak, P., and Laback, B. (2010). Median-plane sound localization as a function of the number of spectral channels using a channel vocoder, J. Acoust. Soc. Am. 127, 990-1001.

Objective:

The dependency of perceived loudness from electrical current in Cochlear Implant (CI) stimulation has been investigated in several existing studies. This investigation has two main goals:

  1. To study the efficiency of an adaptive method to determine the loudness function.
  2. To measure the loudness function in binaural as well as monaural stimulation.

Method:

Loudness functions are measured at single electrodes (or interaural electrode pairs) using the method of categorical loudness scaling. The efficiency of this method for hearing impaired listeners has been demonstrated in previous studies (Brand and Hohmann, JASA 112, p.1597-1604). Both an adaptive method and the method of constant stimuli are used. Binaural functions are measured subsequently to monaural function, including monaural measurements as control conditions.

Application:

The results indicate the suitability and efficiency of the adaptive categorical loudness scaling method as a tool for the fast determination of the loudness function. This can be applied to the clinical fitting of implant processors as well as for pre-measurements in psychoaoustic CI studies. The measurement results also provide new insights into monaural and binaural loudness perception of CI listeners.

Funding:

internal

Publications:

  • Wippel, F., Majdak, P., and Laback, B. (2007). Monaural and binaural categorical loudness scaling in electric hearing, presented at Conference on Implantable Auditory Prostheses (CIAP), Lake Tahoe.
  • Wippel, F. (2007). Monaural and binaural loudness scaling with cochlea implant listeners, master thesis, Technical University Vienna, Autrian Academy of Sciences (in German)

Objective and Methods:

Spectral peaks and notches are important cues that normal hearing listeners use to localize sounds in the vertical planes (the front/back and up/down dimensions). This study investigates to what extent cochlear implant (CI) listeners are sensitive to spectral peaks and notches imposed upon a constant-loudness background. 

Results:

Listeners could always detect peaks, but not always notches. Increasing the bandwidth beyond two electrodes showed no improvement in thresholds. The high-frequency place was significantly worse than the low and middle places; although, listeners had highly-individual tendencies. Thresholds decreased with an increase in the height of the peak. Thresholds for detecting a change in the frequency of a peak or notch were approximately one electrode. Level roving significantly increased thresholds. Thus, there is currently no indication that CI listeners can perform a "true" profile analysis. Future studies will explore if adding temporal cues or roving the level in equal loudness steps, instead of equal-current steps (as in the present study), is relevant for profile analysis.

Application:

Data on the sensitivity to spectral peaks and notches are required to encode spectral localization cues in future CI stimulation strategies. 

Funding:

FWF (Austrian Science Fund): Project #P18401-B15

Publications:

  • Goupell, M., Laback, B., Majdak, P., and Baumgartner, W. D. (2008). Current-level discrimination and spectral profile analysis in multi-channel electrical stimulation, J. Acoust. Soc. Am. 124, 3142-57.
  • Goupell, M. J., Laback, B., Majdak, P., and Baumgartner, W-D. (2007). Sensitivity to spectral peaks and notches in cochlear implant listeners, presented at Conference on Implantable Auditory Prostheses (CIAP), Lake Tahoe.

Objective:

In this project, head-related transfer functions (HRTFs) are measured and prepared for localization tests with cochlear implant listeners. The method and apparatus used for the measurement is the same as used for the general HRTF measurement (see project HRTF-System); however, the place where sound is acquired is different. In this project, the microphones built into the behind-the-ear (BtE) processors of cochlear implantees are used. The processors are located on the pinna, and the unprocessed microphone signals are used to calculate the BtE-HRTFs for different spatial positions.

The BtE-HRTFs are then used in localization tests like Loca BtE-CI.

Objective and Method:

Current cochlear implant (CI) systems are not designed for sound localization in the sagittal planes (front-back and up/down-dimensions). Nevertheless, some of the spectral cues that are important for sagittal plane localization in normal hearing (NH) listeners might be audible for CI listeners. Here, we studied 3-D localization with bilateral CI-listeners using "clinical" CI systems and with NH listeners. Noise sources were filtered with subject-specific head-related transfer functions, and a virtually structured environment was presented via a head-mounted display to provide feedback for learning. 

Results:

The CI listeners performed generally worse than NH listeners, both in the horizontal and vertical dimensions. The localization error decreases with an increase in the duration of training. The front/back confusion rate of trained CI listeners was comparable to that of untrained (naive) NH listeners and two times higher than for the trained NH listeners. 

Application:

The results indicate that some spectral localization cues are available to bilateral CI listeners, even though the localization performance is much worse than for NH listeners. These results clearly show the need for new strategies to encode spectral localization cues for CI listeners, and thus improve sagittal plane localization. Front-back discrimination is particularly important in traffic situations.

Funding:

FWF (Austrian Science Fund): Project # P18401-B15

Publications:

  • Majdak, P., Goupell, M., and Laback, B. (2011). Two-Dimensional Localization of Virtual Sound Sources in Cochlear-Implant Listeners, Ear & Hearing.
  • Majdak, P., Laback, B., and Goupell, M. (2008). 3D-localization of virtual sound sources in normal-hearing and cochlear-implant listeners, presented at Acoustics '08  (ASA-EAA joint) conference, Paris

Objective:

Bilateral cochlear implant (CI) listeners currently use stimulation strategies that encode

interaural time differences (ITD) in the temporal envelope. However, the strategies do not transmit ITD in the fine structure, because of the constant phase in the electric pulse train. To determine the utility of encoding ITD in the fine structure, ITD-based lateralization was investigated with four CI listeners and four normal hearing (NH) subjects who listened to a simulation of electric stimulation.

Methods und Results:

Lateralization discrimination was tested at different pulse rates for various combinations of

independently controlled fine structure ITD and envelope ITD. Results for electric hearing show that the fine structure ITD had the strongest impact on lateralization at lower pulse rates, with significant effects for pulse rates up to 800 pulses per second. At higher pulse rates, lateralization discrimination depended solely on the envelope ITD. The data suggest that bilateral CI listeners benefit from transmitting fine structure ITD at lower pulse rates. However, there were strong inter-individual differences: the better performing CI listeners performed comparably to the NH listeners.

Application:

The result that bilateral CI listeners benefit from transmitting fine structure ITD at lower pulse rates is relevant to future CI stimulation strategies that encode fine timing cues. It is expected that appropriate encoding of these cues improves sound localization abilities and speech understanding in noise.

Funding:

Internal

Publications:

Majdak, P., Laback, B., Baumgartner., W.D. (2006). Effects of interaural time differences in fine structure and envelope on lateral discrimination in electrical hearing, J. Acoust. Soc. Am. 120, 2190-201.

Objective:

This project investigated the perception of interaural intensity differences among cochlear implant (CI) listeners in relation to the spectral composition and the temporal structure of the signal.

Method:

The perception thresholds (just noticeable differences, JND) of CI listeners were examined using differently structured signals. The stimuli were applied directly to the clinical signal processing units, while the parameters of the ongoing stimulation were closely monitored.

Results:

JNDs of IIDs in CI listeners ranged from 1.5 - 2.5 dB for a detection level of 80 percent. The type of stimulus seems to bear little relevance on the detection performance, with the exception of one single type of signal - a pulse train with a frequency of 20 Hz. This means that JNDs of CI listeners are only irrelevantly higher than those of normal hearing listeners. CI implantees are sensitive to IIDs, and the JNDs correlate to a difference in arrival angles ranging from 5-10 degrees. Since the JNDs are within the minimal level widths of the transfer of amplitudes by the CI system, the reduction of level width in future systems seems advisable.

Publication:

  • Laback, B., Pok, S. M., Baumgartner, W. D., Deutsch, W. A., and Schmid, K. (2004). “Sensitivity to interaural level and envelope time differences of two bilateral cochlear implant listeners using clinical sound processors,” Ear and Hearing 25, 5, 488-500.

Objective:

Previous studies show that cochlear implant (CI) listeners show sensitivity to interaural time difference (ITD) in the fine structure at comparable, or sometimes even higher pulse rates than normal hearing (NH) subjects. This study investigates whether the differences between the two subject groups are due to an effect of auditory filtering that is absent in the case of electric stimulation.

Method:

The effects of center frequency and pulse rate on the sensitivity to ongoing envelope ITD were investigated using bandpass-filtered pulse trains. Three center frequencies (4.6, 6.5, and 9.2 kHz) were tested, and the bandwidth was scaled to stimulate an approximately constant range on the basilar membrane. The pulse rate was varied from 200 to 588 pulses per second (pps).

Results:

The results show a small but significant decrease in performance with an increase in center frequency. Furthermore, performance decreases with an increase in pulse rate, yielding a rate limit of approximately 500 pps. The lack of an interaction between pulse rate and center frequency indicates that auditory filtering was not the limiting factor in ITD perception. This suggests the existence of other limiting mechanisms, such as phase locking or more central binaural processes. The comparison of the ITD rate limits in CI subjects with those in NH subjects was considered unaffected by the auditory filtering in NH listeners. 

Funding:

FWF (Austrian Science Fund): Project # P18401-B15

Publications:

  • Laback, B. and Majdak, P. (2007). Effect of Center Frequency on the Sensitivity to Interaural Time Differences in Filtered Pulse Trains, proceedings of DAGA 2007, Stuttgart.
  • Majdak, P., and Laback, B. (2008). Effect of center frequency and rate on the sensitivity to interaural delay in high-frequency click trains, J. Acoust. Soc. Am. (under review).