How neuroscience relates to hearing aid amplification

Comparison of Two-Talker Attention Decoding from EEG with Nonlinear Neural Networks and Linear Methods

Auditory attention decoding (AAD) through a brain-computer interface has had a flowering of developments since it was first introduced by Mesgarani and Chang (2012) using electrocorticograph recordings. AAD has been pursued for its potential application to hearing-aid design in which an attention-guided algorithm selects, from multiple competing acoustic sources, which should be enhanced for the listener and which should be suppressed. Traditionally, researchers have separated the AAD problem into two stages: reconstruction of a representation of the attended audio from neural signals, followed by determining the similarity between the candidate audio streams and the reconstruction. Here, we compare the traditional two-stage approach with a novel neural-network architecture that subsumes the explicit similarity step. We compare this new architecture against linear and non-linear (neural-network) baselines using both wet and dry electroencephalogram (EEG) systems. Our results indicate that the new architecture outperforms the baseline linear stimulus-reconstruction method, improving decoding accuracy from 66% to 81% using wet EEG and from 59% to 87% for dry EEG. Also of note was the finding that the dry EEG system can deliver comparable or even better results than the wet, despite the latter having one third as many EEG channels as the former. The 11-subject, wet-electrode AAD dataset for two competing, co-located talkers, the 11-subject, dry-electrode AAD dataset, and our software are available for further validation, experimentation, and modification.

Listening effort: Are we measuring cognition or affect, or both?

Listening effort is increasingly recognized as a factor in communication, particularly for and with nonnative speakers, for the elderly, for individuals with hearing impairment and/or for those working in noise. However, as highlighted by McGarrigle et al., International Journal of Audiology, 2014, 53, 433-445, the term "listening effort" encompasses a wide variety of concepts, including the engagement and control of multiple possibly distinct neural systems for information processing, and the affective response to the expenditure of those resources in a given context. Thus, experimental or clinical methods intended to objectively quantify listening effort may ultimately reflect a complex interaction between the operations of one or more of those information processing systems, and/or the affective and motivational response to the demand on those systems. Here we examine theoretical, behavioral, and psychophysiological factors related to resolving the question of what we are measuring, and why, when we measure "listening effort." This article is categorized under: Linguistics > Language in Mind and Brain Psychology > Theory and Methods Psychology > Attention Psychology > Emotion and Motivation.

Why Do Hearing Aids Fail to Restore Normal Auditory Perception?

Hearing loss is a widespread condition that is linked to declines in quality of life and mental health. Hearing aids remain the treatment of choice, but, unfortunately, even state-of-the-art devices provide only limited benefit for the perception of speech in noisy environments. While traditionally viewed primarily as a loss of sensitivity, hearing loss is also known to cause complex distortions of sound-evoked neural activity that cannot be corrected by amplification alone. This Opinion article describes the effects of hearing loss on neural activity to illustrate the reasons why current hearing aids are insufficient and to motivate the use of new technologies to explore directions for improving the next generation of devices.

A potential neurophysiological correlate of electric-acoustic pitch matching in adult cochlear implant users: Pilot data

The overall goal of this study was to identify an objective physiological correlate of electric-acoustic pitch matching in unilaterally implanted cochlear implant (CI) participants with residual hearing in the non-implanted ear. Electrical and acoustic stimuli were presented in a continuously alternating fashion across ears. The acoustic stimulus and the electrical stimulus were either matched or mismatched in pitch. Auditory evoked potentials were obtained from nine CI users. Results indicated that N1 latency was stimulus-dependent, decreasing when the acoustic frequency of the tone presented to the non-implanted ear was increased. More importantly, there was an additional decrease in N1 latency in the pitch-matched condition. These results indicate the potential utility of N1 latency as an index of pitch matching in CI users.

Relationship between speech-evoked neural responses and perception of speech in noise in older adults

Speech-in-noise (SPIN) perception involves neural encoding of temporal acoustic cues. Cues include temporal fine structure (TFS) and envelopes that modulate at syllable (Slow-rate ENV) and fundamental frequency (F₀-rate ENV) rates. Here the relationship between speech-evoked neural responses to these cues and SPIN perception was investigated in older adults. Theta-band phase-locking values (PLVs) that reflect cortical sensitivity to Slow-rate ENV and peripheral/brainstem frequency-following responses phase-locked to F₀-rate ENV (FFR_{ENV_F0}) and TFS (FFR_TFS) were measured from scalp-electroencephalography responses to a repeated speech syllable in steady-state speech-shaped noise (SpN) and 16-speaker babble noise (BbN). The results showed that (1) SPIN performance and PLVs were significantly higher under SpN than BbN, implying differential cortical encoding may serve as the neural mechanism of SPIN performance that varies as a function of noise types; (2) PLVs and FFR_TFS at resolved harmonics were significantly related to good SPIN performance, supporting the importance of phase-locked neural encoding of Slow-rate ENV and TFS of resolved harmonics during SPIN perception; (3) FFR_{ENV_F0} was not associated to SPIN performance until audiometric threshold was controlled for, indicating that hearing loss should be carefully controlled when studying the role of neural encoding of F₀-rate ENV. Implications are drawn with respect to fitting auditory prostheses.

Objective assessment of listening effort: effects of an increased task demand

In demanding listening situations the individual has to exert an increased listening effort to process interesting auditory signals correctly. Especially people with hearing loss are particularly affected and require more effort to identify sounds compared to those with normal hearing. So far, a suitable objective estimate of listening effort is still not available. In previous studies, we presented an objective estimate of listening effort (OLEosc), which is based on the instantaneous phase distribution of the ongoing EEG activity. In the current study, the proposed measure was analyzed in detail. The task performance was taken into account, which indicates if the participants can solve the auditory task and exert effort or if they ceased solving the task. The EEG data was acquired from people with moderate hearing loss solving a listening task. During the experiment, hearing aid settings with different microphone configurations were tested. The results indicate that the OLEosc reflects the real exerted effort of the participants. Besides the reflection of different task demands related to the hearing aid settings, it was possible to show a decline of focused attention to the auditory stimuli related to an excessive task demand. Furthermore, the data indicate that the OLEosc is not directly correlated to the speech intelligibility presented in the word score data and is not subjectively biased like the results of the presented rating scale.

Output signal-to-noise ratio and speech perception in noise: effects of algorithm

OBJECTIVE

The aims of this study were to: 1) quantify the amount of change in signal-to-noise ratio (SNR) as a result of compression and noise reduction (NR) processing in devices from three hearing aid (HA) manufacturers and 2) use the SNR changes to predict changes in speech perception. We hypothesised that the SNR change would differ across processing type and manufacturer, and that improvements in SNR would relate to improvements in performance.

DESIGN

SNR at the output of the HAs was quantified using a phase-inversion technique. A linear mixed model was used to determine whether changes in SNR across HA conditions were predictive of changes in aided speech perception in noise.

STUDY SAMPLE

Two groups participated: 25 participants had normal-hearing and 25 participants had mild to moderately severe sensorineural hearing loss.

RESULTS

The HAs programmed for both groups changed the SNR by a small, but statistically significant amount. Significant interactions in SNR changes were observed between HA devices and processing types. However, the change in SNR was not predictive of changes in speech perception.

CONCLUSION

Although observed significant changes in SNR resulting from compression and NR did not convert to changes in speech perception, these algorithms may serve other purposes.

Aging and Hearing Health: The Life-course Approach

Sensory abilities decline with age. More than 5% of the world's population, approximately 360 million people, have disabling hearing loss. In adults, disabling hearing loss is defined by thresholds greater than 40 dBHL in the better hearing ear.Hearing disability is an important issue in geriatric medicine because it is associated with numerous health issues, including accelerated cognitive decline, depression, increased risk of dementia, poorer balance, falls, hospitalizations, and early mortality. There are also social implications, such as reduced communication function, social isolation, loss of autonomy, impaired driving ability, and financial decline. Furthermore, the onset of hearing loss is gradual and subtle, first affecting the detection of high-pitched sounds and with difficulty understanding speech in noisy but not in quiet environments. Consequently, delays in recognizing and seeking help for hearing difficulties are common. Age-related hearing loss has no known cure, and technologies (hearing aids, cochlear implants, and assistive devices) improve thresholds but do not restore hearing to normal. Therefore, health care for persons with hearing loss and people within their communication circles requires education and counseling (e.g., increasing knowledge, changing attitudes, and reducing stigma), behavior change (e.g., adapting communication strategies), and environmental modifications (e.g., reducing noise). In this article, we consider the causes, consequences, and magnitude of hearing loss from a life-course perspective. We examine the concept of "hearing health," how to achieve it, and implications for policy and practice.

Listening and Learning: Cognitive Contributions to the Rehabilitation of Older Adults With and Without Audiometrically Defined Hearing Loss

Here, we describe some of the ways in which aging negatively affects the way sensory input is transduced and processed within the aging brain and how cognitive work is involved when listening to a less-than-perfect signal. We also describe how audiologic rehabilitation, including hearing aid amplification and listening training, is used to reduce the amount of cognitive resources required for effective auditory communication and conclude with an example of how listening effort is being studied in research laboratories for the purpose(s) of informing clinical practice.

Aided Electrophysiology Using Direct Audio Input: Effects of Amplification and Absolute Signal Level

PURPOSE

This study investigated (a) the effect of amplification on cortical auditory evoked potentials (CAEPs) at different signal levels when signal-to-noise ratios (SNRs) were equated between unaided and aided conditions, and (b) the effect of absolute signal level on aided CAEPs when SNR was held constant.

METHOD

CAEPs were recorded from 13 young adults with normal hearing. A 1000-Hz pure tone was presented in unaided and aided conditions with a linear analog hearing aid. Direct audio input was used, allowing recorded hearing aid noise floor to be added to unaided conditions to equate SNRs between conditions. An additional stimulus was created through scaling the noise floor to study the effect of signal level.

RESULTS

Amplification resulted in delayed N1 and P2 peak latencies relative to the unaided condition. An effect of absolute signal level (when SNR was constant) was present for aided CAEP area measures, such that larger area measures were found at higher levels.

CONCLUSION

Results of this study further demonstrate that factors in addition to SNR must also be considered before CAEPs can be used to clinically to measure aided thresholds.

The neural encoding of formant frequencies contributing to vowel identification in normal-hearing listeners

Even though speech signals trigger coding in the cochlea to convey speech information to the central auditory structures, little is known about the neural mechanisms involved in such processes. The purpose of this study was to understand the encoding of formant cues and how it relates to vowel recognition in listeners. Neural representations of formants may differ across listeners; however, it was hypothesized that neural patterns could still predict vowel recognition. To test the hypothesis, the frequency-following response (FFR) and vowel recognition were obtained from 38 normal-hearing listeners using four different vowels, allowing direct comparisons between behavioral and neural data in the same individuals. FFR was employed because it provides an objective and physiological measure of neural activity that can reflect formant encoding. A mathematical model was used to describe vowel confusion patterns based on the neural responses to vowel formant cues. The major findings were (1) there were large variations in the accuracy of vowel formant encoding across listeners as indexed by the FFR, (2) these variations were systematically related to vowel recognition performance, and (3) the mathematical model of vowel identification was successful in predicting good vs poor vowel identification performers based exclusively on physiological data.

The Ear-Brain Connection: Older Ears and Older Brains

PURPOSE

The purpose of this article is to review recent research from our laboratory on the topic of aging, and the ear-brain system, as it relates to hearing aid use and auditory rehabilitation. The material described here was presented as part of the forum on the brain and hearing aids, at the 2014 HEaling Across the Lifespan (HEAL) conference.

METHOD

The method involves a narrative review of previously reported electroencephalography (EEG) and magnetoencephalography (MEG) data from our laboratory as they relate to the (a) neural detection of amplified sound and (b) ability to learn new sound contrasts.

CONCLUSIONS

Results from our studies add to the mounting evidence that there are central effects of biological aging as well as peripheral pathology that affect a person's neural detection and use of sound. What is more, these biological effects can be seen as early as middle age. The accruing evidence has implications for hearing aid use because effective communication relies not only on sufficient detection of sound but also on the individual's ability to learn to make use of these sounds in ever-changing listening environments.