Objective detection of evoked potentials using a bootstrap technique

Evaluating Fixed Single-Point Parameters When Applied to Vestibular Evoked Myogenic Potentials: The Effect of Single Point and Signal Window

OBJECTIVES

Several studies have applied a common objective detection algorithm (fixed single point [ Fsp ]) for detection of the vestibular evoked myogenic potential (VEMP). However, fundamental parameters of Fsp , such as establishing the location and duration of a signal window, have not been examined. In addition, Fsp criterion values used for response detection have not been established for cervical VEMPs (cVEMPs) or ocular VEMPs (oVEMPs). The purpose of this article was to investigate the effect of various single points and signal windows on Fsp , as well as determining Fsp criteria to determine response presence for cVEMP and oVEMP in a group of young healthy participants.

DESIGN

Twenty young healthy adults under the age of 30 and with no history of hearing or balance concerns were enrolled in the study protocol. Air-conducted cVEMPs and oVEMPs were evoked using 500 Hz tone bursts at 123 dB pSPL recorded at a fixed electromyography activation of 50 µV for cVEMPs and 35° gaze angle for oVEMPs. Responses were analyzed off-line using visual and objective detection. Fsp was applied to cVEMPs and oVEMPs using a range of single points and signal windows.

RESULTS

Noise variance was lowest for cVEMPs at the latency of P1, and for oVEMPs noise variance was not significantly different across the single-point latencies. On average, extending the length of the signal window lowered the Fsp value in cVEMPs and oVEMPs. An Fsp value of 2.0 was chosen as the criterion cutoff associated with the 95th percentile during no-response conditions using group data for cVEMPs and oVEMPs, respectively. Fsp values for cVEMPs and oVEMPs were not significantly different from each other.

DISCUSSION

This study established single-point latency and time-window parameters for VEMP-related applications of the Fsp detection algorithm. Fsp criteria values were established for cVEMP and oVEMP. Using these parameters, responses were detected in all participants.

Detecting cortical responses to continuous running speech using EEG data from only one channel

OBJECTIVE

To explore the detection of cortical responses to continuous speech using a single EEG channel. Particularly, to compare detection rates and times using a cross-correlation approach and parameters extracted from the temporal response function (TRF).

DESIGN

EEG from 32-channels were recorded whilst presenting 25-min continuous English speech. Detection parameters were cross-correlation between speech and EEG (XCOR), peak value and power of the TRF filter (TRF-peak and TRF-power), and correlation between predicted TRF and true EEG (TRF-COR). A bootstrap analysis was used to determine response statistical significance. Different electrode configurations were compared: Using single channels Cz or Fz, or selecting channels with the highest correlation value.

STUDY SAMPLE

Seventeen native English-speaking subjects with mild-to-moderate hearing loss.

RESULTS

Significant cortical responses were detected from all subjects at Fz channel with XCOR and TRF-COR. Lower detection time was seen for XCOR (mean = 4.8 min) over TRF parameters (best TRF-COR, mean = 6.4 min), with significant time differences from XCOR to TRF-peak and TRF-power. Analysing multiple EEG channels and testing channels with the highest correlation between envelope and EEG reduced detection sensitivity compared to Fz alone.

CONCLUSIONS

Cortical responses to continuous speech can be detected from a single channel with recording times that may be suitable for clinical application.

Modified T2 Statistics for Improved Detection of Aided Cortical Auditory Evoked Potentials in Hearing-Impaired Infants

The cortical auditory evoked potential (CAEP) is a change in neural activity in response to sound, and is of interest for audiological assessment of infants, especially those who use hearing aids. Within this population, CAEP waveforms are known to vary substantially across individuals, which makes detecting the CAEP through visual inspection a challenging task. It also means that some of the best automated CAEP detection methods used in adults are probably not suitable for this population. This study therefore evaluates and optimizes the performance of new and existing methods for aided (i.e., the stimuli are presented through subjects' hearing aid(s)) CAEP detection in infants with hearing loss. Methods include the conventional Hotellings T2 test, various modified q-sample statistics, and two novel variants of T2 statistics, which were designed to exploit the correlation structure underlying the data. Various additional methods from the literature were also evaluated, including the previously best-performing methods for adult CAEP detection. Data for the assessment consisted of aided CAEPs recorded from 59 infant hearing aid users with mild to profound bilateral hearing loss, and simulated signals. The highest test sensitivities were observed for the modified T2 statistics, followed by the modified q-sample statistics, and lastly by the conventional Hotelling's T2 test, which showed low detection rates for ensemble sizes <80 epochs. The high test sensitivities at small ensemble sizes observed for the modified T2 and q-sample statistics are especially relevant for infant testing, as the time available for data collection tends to be limited in this population.

Objective hearing threshold identification from auditory brainstem response measurements using supervised and self-supervised approaches

Hearing loss is a major health problem and psychological burden in humans. Mouse models offer a possibility to elucidate genes involved in the underlying developmental and pathophysiological mechanisms of hearing impairment. To this end, large-scale mouse phenotyping programs include auditory phenotyping of single-gene knockout mouse lines. Using the auditory brainstem response (ABR) procedure, the German Mouse Clinic and similar facilities worldwide have produced large, uniform data sets of averaged ABR raw data of mutant and wildtype mice. In the course of standard ABR analysis, hearing thresholds are assessed visually by trained staff from series of signal curves of increasing sound pressure level. This is time-consuming and prone to be biased by the reader as well as the graphical display quality and scale.In an attempt to reduce workload and improve quality and reproducibility, we developed and compared two methods for automated hearing threshold identification from averaged ABR raw data: a supervised approach involving two combined neural networks trained on human-generated labels and a self-supervised approach, which exploits the signal power spectrum and combines random forest sound level estimation with a piece-wise curve fitting algorithm for threshold finding.We show that both models work well and are suitable for fast, reliable, and unbiased hearing threshold detection and quality control. In a high-throughput mouse phenotyping environment, both methods perform well as part of an automated end-to-end screening pipeline to detect candidate genes for hearing involvement. Code for both models as well as data used for this work are freely available.

On prediction of aided behavioural measures using speech auditory brainstem responses and decision trees

Current clinical strategies to assess benefits from hearing aids (HAs) are based on self-reported questionnaires and speech-in-noise (SIN) tests; which require behavioural cooperation. Instead, objective measures based on Auditory Brainstem Responses (ABRs) to speech stimuli would not require the individuals' cooperation. Here, we re-analysed an existing dataset to predict behavioural measures with speech-ABRs using regression trees. Ninety-two HA users completed a self-reported questionnaire (SSQ-Speech) and performed two aided SIN tests: sentences in noise (BKB-SIN) and vowel-consonant-vowels (VCV) in noise. Speech-ABRs were evoked by a 40 ms [da] and recorded in 2x2 conditions: aided vs. unaided and quiet vs. background noise. For each recording condition, two sets of features were extracted: 1) amplitudes and latencies of speech-ABR peaks, 2) amplitudes and latencies of speech-ABR F0 encoding. Two regression trees were fitted for each of the three behavioural measures with either feature set and age, digit-span forward and backward, and pure tone average (PTA) as possible predictors. The PTA was the only predictor in the SSQ-Speech trees. In the BKB-SIN trees, performance was predicted by the aided latency of peak F in quiet for participants with PTAs between 43 and 61 dB HL. In the VCV trees, performance was predicted by the aided F0 encoding latency and the aided amplitude of peak VA in quiet for participants with PTAs ≤ 47 dB HL. These findings indicate that PTA was more informative than any speech-ABR measure, as these were relevant only for a subset of the participants. Therefore, speech-ABRs evoked by a 40 ms [da] are not a clinical predictor of behavioural measures in HA users.

Auditory Brainstem Response Detection Using Machine Learning: A Comparison With Statistical Detection Methods

OBJECTIVES

The primary objective of this study was to train and test machine learning algorithms to be able to detect accurately whether EEG data contains an auditory brainstem response (ABR) or not and recommend suitable machine learning methods. In addition, the performance of the best machine learning algorithm was compared with that of prominent statistical detection methods.

DESIGN

Four machine learning algorithms were trained and evaluated using nested k-fold cross-validation: a random forest, a convolutional long short-term memory network, a stacked ensemble, and a multilayer perceptron. The best method was evaluated on a separate test set and compared with conventional detection methods: Fsp, Fmp, q-sample uniform scores test, and Hotelling's T2 test. The models were trained and tested on simulated data that were generated based on recorded ABRs collected from 12 normal-hearing participants and no-stimulus EEG data from 15 participants. Simulation allowed the ground truth of the data ("response present" or "response absent") to be known.

RESULTS

The sensitivity of the best machine learning algorithm, a stacked ensemble, was significantly greater than that of the conventional detection methods evaluated. The stacked ensemble, evaluated using a bootstrap approach, consistently achieved a high and stable level of specificity across ensemble sizes.

CONCLUSIONS

The stacked ensemble model presented was more effective than conventional statistical ABR detection methods and the alternative machine learning approaches tested. The stacked ensemble detection method may have potential both in automated ABR screening devices as well as in evoked potential software, assisting clinicians in making decisions regarding a patient's ABR threshold. Further assessment of the model's generalizability using a large cohort of subject recorded data, including participants of different ages and hearing status, is a recommended next step.

Controlling test specificity for auditory evoked response detection using a frequency domain bootstrap

BACKGROUND

Statistical detection methods are routinely used to automate auditory evoked response (AER) detection and assist clinicians with AER measurements. However, many of these methods are built around statistical assumptions that can be violated for AER data, potentially resulting in reduced or unpredictable test performances. This study explores a frequency domain bootstrap (FDB) and some FDB modifications to preserve test performance in serially correlated non-stationary data.

METHOD

The FDB aims to generate many surrogate recordings, all with similar serial correlation as the original recording being analysed. Analysing the surrogates with the detection method then gives a distribution of values that can be used for inference. A potential limitation of the conventional FDB is the assumption of stationary data with a smooth power spectral density (PSD) function, which is addressed through two modifications.

COMPARISONS WITH EXISTING METHODS

The FDB was compared to a conventional parametric approach and two modified FDB approaches that aim to account for heteroskedasticity and non-smooth PSD functions. Hotelling's T²(HT2) test applied to auditory brainstem responses was the test case.

RESULTS

When using conventional HT2, false-positive rates deviated significantly from the nominal alpha-levels due to serial correlation. The false-positive rates of the modified FDB were consistently closer to the nominal alpha-levels, especially when data was strongly heteroskedastic or the underlying PSD function was not smooth due to e.g. power lines noise.

CONCLUSION

The FDB and its modifications provide accurate, recording-dependent approximations of null distributions, and an improved control of false-positive rates relative to parametric inference for auditory brainstem response detection.

Hearing sensitivity: An underlying mechanism for niche differentiation in gleaning bats

Tropical ecosystems are known for high species diversity. Adaptations permitting niche differentiation enable species to coexist. Historically, research focused primarily on morphological and behavioral adaptations for foraging, roosting, and other basic ecological factors. Another important factor, however, is differences in sensory capabilities. So far, studies mainly have focused on the output of behavioral strategies of predators and their prey preference. Understanding the coexistence of different foraging strategies, however, requires understanding underlying cognitive and neural mechanisms. In this study, we investigate hearing in bats and how it shapes bat species coexistence. We present the hearing thresholds and echolocation calls of 12 different gleaning bats from the ecologically diverse Phyllostomid family. We measured their auditory brainstem responses to assess their hearing sensitivity. The audiograms of these species had similar overall shapes but differed substantially for frequencies below 9 kHz and in the frequency range of their echolocation calls. Our results suggest that differences among bats in hearing abilities contribute to the diversity in foraging strategies of gleaning bats. We argue that differences in auditory sensitivity could be important mechanisms shaping diversity in sensory niches and coexistence of species.

Perturbation of resting-state network nodes preferentially propagates to structurally rather than functionally connected regions

Combining Transcranial Magnetic Stimulation (TMS) with electroencephalography (EEG) offers the opportunity to study signal propagation dynamics at high temporal resolution in the human brain. TMS pulse induces a local effect which propagates across cortical networks engaging distant cortical and subcortical sites. However, the degree of propagation supported by the structural compared to functional connectome remains unclear. Clarifying this issue would help tailor TMS interventions to maximize target engagement. The goal of this study was to establish the contribution of functional and structural connectivity in predicting TMSinduced signal propagation after perturbation of two distinct brain networks. For this purpose, 24 healthy individuals underwent two identical TMS-EEG visits where neuronavigated TMS pulses were delivered to nodes of the default mode network (DMN) and the dorsal attention network (DAN). The functional and structural connectivity derived from each individual stimulation spot were characterized via functional magnetic resonance imaging (fMRI) and Diffusion Weighted Imaging (DWI), and signal propagation across these two metrics was compared. Direct comparison between the signal extracted from brain regions either functionally or structurally connected to the stimulation sites, shows a stronger activation over cortical areas connected via white matter pathways, with a minor contribution of functional projections. This pattern was not observed when analyzing spontaneous resting state EEG activity. Overall, results suggest that structural links can predict network-level response to perturbation more accurately than functional connectivity. Additionally, DWI-based estimation of propagation patterns can be used to estimate off-target engagement of other networks and possibly guide target selection to maximize specificity.

Network-level macroscale structural connectivity predicts propagation of transcranial magnetic stimulation

Information processing in the brain is mediated by structural white matter pathways and is highly dependent on topological brain properties. Here we combined transcranial magnetic stimulation (TMS) with high-density electroencephalography (EEG) and Diffusion Weighted Imaging (DWI), specifically looking at macroscale connectivity to understand whether regional, network-level or whole-brain structural properties are more responsible for stimulus propagation. Neuronavigated TMS pulses were delivered over two individually defined nodes of the default mode (DMN) and dorsal attention (DAN) networks in a group of healthy subjects, with test-retest reliability assessed 1-month apart. TMS-evoked activity was predicted by the modularity and structural integrity of the stimulated network rather than the targeted region(s) or the whole-brain connectivity, suggesting network-level structural connectivity as more relevant than local and global brain properties in shaping TMS signal propagation. The importance of network structural connectome was unveiled only by evoked activity, but not resting-state data. Future clinicals interventions might enhance target engagement by adopting DWI-guided, network-focused TMS.

Hearing sensitivity and amplitude coding in bats are differentially shaped by echolocation calls and social calls

Differences in auditory perception between species are influenced by phylogenetic origin and the perceptual challenges imposed by the natural environment, such as detecting prey- or predator-generated sounds and communication signals. Bats are well suited for comparative studies on auditory perception since they predominantly rely on echolocation to perceive the world, while their social calls and most environmental sounds have low frequencies. We tested if hearing sensitivity and stimulus level coding in bats differ between high and low-frequency ranges by measuring auditory brainstem responses (ABRs) of 86 bats belonging to 11 species. In most species, auditory sensitivity was equally good at both high- and low-frequency ranges, while amplitude was more finely coded for higher frequency ranges. Additionally, we conducted a phylogenetic comparative analysis by combining our ABR data with published data on 27 species. Species-specific peaks in hearing sensitivity correlated with peak frequencies of echolocation calls and pup isolation calls, suggesting that changes in hearing sensitivity evolved in response to frequency changes of echolocation and social calls. Overall, our study provides the most comprehensive comparative assessment of bat hearing capacities to date and highlights the evolutionary pressures acting on their sensory perception.

Efficient Detection of Cortical Auditory Evoked Potentials in Adults Using Bootstrapped Methods

BACKGROUND

Statistical detection methods are useful tools for assisting clinicians with cortical auditory evoked potential (CAEP) detection, and can help improve the overall efficiency and reliability of the test. However, many of these detection methods rely on parametric distributions when evaluating test significance, and thus make various assumptions regarding the electroencephalogram (EEG) data. When these assumptions are violated, reduced test sensitivities and/or increased or decreased false-positive rates can be expected. As an alternative to the parametric approach, test significance can be evaluated using a bootstrap, which does not require some of the aforementioned assumptions. Bootstrapping also permits a large amount of freedom when choosing or designing the statistical test for response detection, as the distributions underlying the test statistic no longer need to be known prior to the test.

OBJECTIVES

To improve the reliability and efficiency of CAEP-related applications by improving the specificity and sensitivity of objective CAEP detection methods.

DESIGN

The methods included in the assessment were Hotelling's T2 test, the Fmp, four modified q-sample statistics, and various template-based detection methods (calculated between the ensemble coherent average and some predefined template), including the correlation coefficient, covariance, and dynamic time-warping (DTW). The assessment was carried out using both simulations and a CAEP threshold series collected from 23 adults with normal hearing.

RESULTS

The most sensitive method was DTW, evaluated using the bootstrap, with maximum increases in test sensitivity (relative to the conventional Hotelling's T2 test) of up to 30%. An important factor underlying the performance of DTW is that the template adopted for the analysis correlates well with the subjects' CAEP.

CONCLUSION

When subjects' CAEP morphology is approximately known before the test, then the DTW algorithm provides a highly sensitive method for CAEP detection.

Ocular and cervical vestibular evoked myogenic potentials elicited by air-conducted, low-frequency sound

BACKGROUND

Sound is not only detected by the cochlea, but also, at high intensities, by the vestibular system. Acoustic activation of the vestibular system can manifest itself in vestibular evoked myogenic potentials (VEMPs). In a clinical setting, VEMPs are usually evoked with rather high-frequency sound (500 Hz and higher), despite the fact that only a fraction of saccular and utricular hair cells in the striolar region is available for high-frequency stimulation.

OBJECTIVE

As a growing proportion of the population complains about low-frequency environmental noise, including reports on vestibular symptoms, the activation of the vestibular system by low-frequency sound deserves better understanding.

METHODS

We recorded growth functions of oVEMPs and cVEMPs evoked with air-conducted sound at 120 Hz and below. We estimated VEMP thresholds and tested whether phase changes of the stimulus carrier result in changes of VEMP amplitude and latency.

RESULTS

The VEMP response of the otholith organs to low-frequency sound is uniform and not tuned when corrected for middle ear attenuation by A-weighting the stimulus level. Different stimulus carrier phases result in phase-correlated changes of cVEMP latencies and amplitudes.

CONCLUSIONS

VEMPs can be evoked with rather low-frequency sound, but high thresholds suggest that they are unlikely to be triggered by environmental sounds.

VEP estimation of visual acuity: a systematic review

Purpose

Visual evoked potentials (VEPs) can be used to measure visual resolution via a spatial frequency (SF) limit as an objective estimate of visual acuity. The aim of this systematic review is to collate descriptions of the VEP SF limit in humans, healthy and disordered, and to assess how accurately and precisely VEP SF limits reflect visual acuity.

Methods

The protocol methodology followed the PRISMA statement. Multiple databases were searched using “VEP” and “acuity” and associated terms, plus hand search: titles, abstracts or full text were reviewed for eligibility. Data extracted included VEP SF limits, stimulus protocols, VEP recording and analysis techniques and correspondence with behavioural acuity for normally sighted healthy adults, typically developing infants and children, healthy adults with artificially degraded vision and patients with ophthalmic or neurological conditions.

Results

A total of 155 studies are included. Commonly used stimulus, recording and analysis techniques are summarised. Average healthy adult VEP SF limits vary from 15 to 40 cpd, depend on stimulus, recording and analysis techniques and are often, but not always, poorer than behavioural acuity measured either psychophysically with an identical stimulus or with a clinical acuity test. The difference between VEP SF limit and behavioural acuity is variable and strongly dependent on the VEP stimulus and choice of acuity test. VEP SF limits mature rapidly, from 1.5 to 9 cpd by the end of the first month of life to 12–20 cpd by 8–12 months, with slower improvement to 20–40 cpd by 3–5 years. VEP SF limits are much better than behavioural thresholds in the youngest, typically developing infants. This difference lessens with age and reaches equivalence between 1 and 2 years; from around 3–5 years, behavioural acuity is better than the VEP SF limit, as for adults. Healthy, artificially blurred adults had slightly better behavioural acuity than VEP SF limits across a wide range of acuities, while adults with heterogeneous ophthalmic or neurological pathologies causing reduced acuity showed a much wider and less consistent relationship. For refractive error, ocular media opacity or pathology primarily affecting the retina, VEP SF limits and behavioural acuity had a fairly consistent relationship across a wide range of acuity. This relationship was much less consistent or close for primarily macular, optic nerve or neurological conditions such as amblyopia. VEP SF limits were almost always normal in patients with non-organic visual acuity loss.

Conclusions

The VEP SF limit has great utility as an objective acuity estimator, especially in pre-verbal children or patients of any age with motor or learning impairments which prevent reliable measurement of behavioural acuity. Its diagnostic power depends heavily on adequate, age-stratified, reference data, age-stratified empirical calibration with behavioural acuity, and interpretation in the light of other electrophysiological and clinical findings. Future developments could encompass faster, more objective and robust techniques such as real-time, adaptive control.

Registration

International prospective register of systematic reviews PROSPERO (https://www.crd.york.ac.uk/PROSPERO/), registration number CRD42018085666.

Post-natal development of the envelope following response to amplitude modulated sounds in the bat Phyllostomus discolor

Bats use a large repertoire of calls for social communication, which are often characterized by temporal amplitude and frequency modulations. As bats are considered to be among the few mammalian species capable of vocal learning, the perception of temporal sound modulations should be crucial for juvenile bats to develop social communication abilities. However, the post-natal development of auditory processing of temporal modulations has not been investigated in bats, so far. Here we use the minimally invasive technique of recording auditory brainstem responses to measure the envelope following response (EFR) to sinusoidally amplitude modulated noise (range of modulation frequencies: 11-130 Hz) in three juveniles (p8-p72) of the bat, Phyllostomus discolor. In two out of three animals, we show that although amplitude modulation processing is basically developed at p8, EFRs maturated further over a period of about two weeks until p33. Maturation of the EFR generally took longer for higher modulation frequencies (87-130 Hz) than for lower modulation frequencies (11-58 Hz).

Speech Auditory Brainstem Responses in Adult Hearing Aid Users: Effects of Aiding and Background Noise, and Prediction of Behavioral Measures

Evaluation of patients who are unable to provide behavioral responses on standard clinical measures is challenging due to the lack of standard objective (non-behavioral) clinical audiological measures that assess the outcome of an intervention (e.g., hearing aids). Brainstem responses to short consonant-vowel stimuli (speech-auditory brainstem responses [speech-ABRs]) have been proposed as a measure of subcortical encoding of speech, speech detection, and speech-in-noise performance in individuals with normal hearing. Here, we investigated the potential application of speech-ABRs as an objective clinical outcome measure of speech detection, speech-in-noise detection and recognition, and self-reported speech understanding in 98 adults with sensorineural hearing loss. We compared aided and unaided speech-ABRs, and speech-ABRs in quiet and in noise. In addition, we evaluated whether speech-ABR F0 encoding (obtained from the complex cross-correlation with the 40 ms [da] fundamental waveform) predicted aided behavioral speech recognition in noise or aided self-reported speech understanding. Results showed that (a) aided speech-ABRs had earlier peak latencies, larger peak amplitudes, and larger F0 encoding amplitudes compared to unaided speech-ABRs; (b) the addition of background noise resulted in later F0 encoding latencies but did not have an effect on peak latencies and amplitudes or on F0 encoding amplitudes; and (c) speech-ABRs were not a significant predictor of any of the behavioral or self-report measures. These results show that speech-ABR F0 encoding is not a good predictor of speech-in-noise recognition or self-reported speech understanding with hearing aids. However, our results suggest that speech-ABRs may have potential for clinical application as an objective measure of speech detection with hearing aids.

Auditory brainstem responses in the bat Carollia perspicillata: threshold calculation and relation to audiograms based on otoacoustic emission measurement

An objective method to evaluate auditory brainstem-evoked responses (ABR) based on the root-mean-square (rms) amplitude of the measured signal and bootstrapping procedures was used to determine threshold curves (see Lv et al. in Med Eng Phys 29:191-198, 2007; Linnenschmidt and Wiegrebe in Hear Res 373:85-95, 2019). The rms values and their significance for threshold determination depended strongly on the filtering of the signal. Using the minimum threshold values obtained at three different low-frequency filter corner frequencies (30, 100, 300 Hz), ABR threshold curves were calculated. The course of the ABR thresholds was comparable to that of published DPOAE (distortion-product otoacoustic emission) thresholds based on a - 10 dB SPL threshold criterion for the 2f1-f2 emission (Schlenther et al. in J Assoc Res Otolaryngol 15:695-705, 2014, frequency range 10-90 kHz). For frequencies between 20 and 80 kHz, which is the most sensitive part of the bat's audiogram, median thresholds ranged between 10 and 28 dB SPL, and the DPOAE thresholds ranged between 10 and 23 dB SPL. At frequencies below 20 kHz (5-20 kHz) and above 80 kHz (80-120 kHz), ABR thresholds increased by 20 dB/octave and 45 dB/octave, respectively. We conclude that the combination of objective threshold determination and multiple filtering of the signal gives reliable ABR thresholds comparable to cochlear threshold curves.

Speech Auditory Brainstem Responses: Effects of Background, Stimulus Duration, Consonant-Vowel, and Number of Epochs

Supplemental Digital Content is available in the text.

Objectives:

The aims of this study were to systematically explore the effects of stimulus duration, background (quiet versus noise), and three consonant–vowels on speech-auditory brainstem responses (ABRs). Additionally, the minimum number of epochs required to record speech-ABRs with clearly identifiable waveform components was assessed. The purpose was to evaluate whether shorter duration stimuli could be reliably used to record speech-ABRs both in quiet and in background noise to the three consonant–vowels, as opposed to longer duration stimuli that are commonly used in the literature. Shorter duration stimuli and a smaller number of epochs would require shorter test sessions and thus encourage the transition of the speech-ABR from research to clinical practice.

Design:

Speech-ABRs in response to 40 msec [da], 50 msec [ba] [da] [ga], and 170 msec [ba] [da] [ga] stimuli were collected from 12 normal-hearing adults with confirmed normal click-ABRs. Monaural (right-ear) speech-ABRs were recorded to all stimuli in quiet and to 40 msec [da], 50 msec [ba] [da] [ga], and 170 msec [da] in a background of two-talker babble at +10 dB signal to noise ratio using a 2-channel electrode montage (Cz-Active, A1 and A2-reference, Fz-ground). Twelve thousand epochs (6000 per polarity) were collected for each stimulus and background from all participants. Latencies and amplitudes of speech-ABR peaks (V, A, D, E, F, O) were compared across backgrounds (quiet and noise) for all stimulus durations, across stimulus durations (50 and 170 msec) and across consonant–vowels ([ba], [da], and [ga]). Additionally, degree of phase locking to the stimulus fundamental frequency (in quiet versus noise) was evaluated for the frequency following response in speech-ABRs to the 170 msec [da]. Finally, the number of epochs required for a robust response was evaluated using F_sp statistic and bootstrap analysis at different epoch iterations.

Results:

Background effect: the addition of background noise resulted in speech-ABRs with longer peak latencies and smaller peak amplitudes compared with speech-ABRs in quiet, irrespective of stimulus duration. However, there was no effect of background noise on the degree of phase locking of the frequency following response to the stimulus fundamental frequency in speech-ABRs to the 170 msec [da]. Duration effect: speech-ABR peak latencies and amplitudes did not differ in response to the 50 and 170 msec stimuli. Consonant–vowel effect: different consonant–vowels did not have an effect on speech-ABR peak latencies regardless of stimulus duration. Number of epochs: a larger number of epochs was required to record speech-ABRs in noise compared with in quiet, and a smaller number of epochs was required to record speech-ABRs to the 40 msec [da] compared with the 170 msec [da].

Conclusions:

This is the first study that systematically investigated the clinical feasibility of speech-ABRs in terms of stimulus duration, background noise, and number of epochs. Speech-ABRs can be reliably recorded to the 40 msec [da] without compromising response quality even when presented in background noise. Because fewer epochs were needed for the 40 msec [da], this would be the optimal stimulus for clinical use. Finally, given that there was no effect of consonant–vowel on speech-ABR peak latencies, there is no evidence that speech-ABRs are suitable for assessing auditory discrimination of the stimuli used.

Objective Comparison of the Quality and Reliability of Auditory Brainstem Response Features Elicited by Click and Speech Sounds

OBJECTIVES

Auditory brainstem responses (ABRs) are commonly generated using simple, transient stimuli (e.g., clicks or tone bursts). While resulting waveforms are undeniably valuable clinical tools, they are unlikely to be representative of responses to more complex, behaviorally relevant sounds such as speech. There has been interest in the use of more complex stimuli to elicit the ABR, with considerable work focusing on the use of synthetically generated consonant-vowel (CV) stimuli. Such responses may be sensitive to a range of clinical conditions and to the effects of auditory training. Several ABR features have been documented in response to CV stimuli; however, an important issue is how robust such features are. In the current research, we use time- and frequency-domain objective measures of quality to compare the reliability of Wave V of the click-evoked ABR to that of waves elicited by the CV stimulus /da/.

DESIGN

Stimuli were presented to 16 subjects at 70 dB nHL in quiet for 6000 epochs. The presence and quality of response features across subjects were examined using Fsp and a Bootstrap analysis method, which was used to assign p values to ABR features for individual recordings in both time and frequency domains.

RESULTS

All consistent peaks identified within the /da/-evoked response had significantly lower amplitude than Wave V of the ABR. The morphology of speech-evoked waveforms varied across subjects. Mean Fsp values for several waves of the speech-evoked ABR were below 3, suggesting low quality. The most robust response to the /da/ stimulus appeared to be an offset response. Only click-evoked Wave V showed 100% wave presence. Responses to the /da/ stimulus showed lower wave detectability. Frequency-domain analysis showed stronger and more consistent activity evoked by clicks than by /da/. Only the click ABR had consistent time-frequency domain features across all subjects.

CONCLUSIONS

Based on the objective analysis used within this investigation, it appears that the quality of speech-evoked ABR is generally less than that of click-evoked responses, although the quality of responses may be improved by increasing the number of epochs or the stimulation level. This may have implications for the clinical use of speech-evoked ABR.

Bistability, Causality, and Complexity in Cortical Networks: An In Vitro Perturbational Study

Measuring the spatiotemporal complexity of cortical responses to direct perturbations provides a reliable index of the brain's capacity for consciousness in humans under both physiological and pathological conditions. Upon loss of consciousness, the complex pattern of causal interactions observed during wakefulness collapses into a stereotypical slow wave, suggesting that cortical bistability may play a role. Bistability is mainly expressed in the form of slow oscillations, a default pattern of activity that emerges from cortical networks in conditions of functional or anatomical disconnection. Here, we employ an in vitro model to understand the relationship between bistability and complexity in cortical circuits. We adapted the perturbational complexity index applied in humans to electrically stimulated cortical slices under different neuromodulatory conditions. At this microscale level, we demonstrate that perturbational complexity can be effectively modulated by pharmacological reduction of bistability and, albeit to a lesser extent, by enhancement of excitability, providing mechanistic insights into the macroscale measurements performed in humans.

A group sequential test for ABR detection

Objective: To detect the auditory brainstem response (ABR) automatically using an innovative sequentially applied Hotelling's T ² test, with the overall goal of optimising test time whilst controlling the false-positive rate (FPR). Design: The stage-wise critical decision boundaries for accepting or rejecting the null hypothesis were found using a new approach called the Convolutional Group Sequential Test (CGST). Specificity, sensitivity, and test time were evaluated using simulations and subject recorded data. Study sample: Data consists of click-evoked ABR threshold series from 12 normal hearing adults, and recordings of EEG background activity from 17 normal hearing adults. Results: Reductions in mean test time of up to 40-45% were observed for the sequential test, relative to a conventional "single shot" test where the statistical test is applied to the data just once. To obtain these results, it will occasionally be necessary to run the test to a higher number of stimuli, i.e. the maximum test time needs to be increased. Conclusions: The CGST can be used to control the specificity of a sequentially applied ABR detection method. Doing so can reduce test time, relative to the "single shot" test, when considered across a cohort of test subjects.

The Convolutional Group Sequential Test: Reducing Test Time for Evoked Potentials

When using a statistical test for automatically detecting evoked potentials, the number of stimuli presented to the subject (the sample size for the statistical test) should be specified at the outset. For evoked response detection, this may be inefficient, i.e., because the signal-to-noise ratio (SNR) of the response is not known in advance, the user would usually err on the cautious side and use a relatively high number of stimuli to ensure adequate statistical power. A more efficient approach is to apply the statistical test repeatedly to the accumulating data over time, as this allows the test to be stopped early for the high SNR responses (thus reducing test time), or later for the low SNR responses. The caveat is that the critical decision boundaries for rejecting the null hypothesis need to be adjusted if the intended type-I error rate is to be obtained. This study presents an intuitive and flexible method for controlling the type-I error rate for sequentially applied statistical tests. The method is built around the discrete convolution of truncated probability density functions, which allows the null distribution for the test statistic to be constructed at each stage of the sequential analysis. Because the null distribution remains tractable, the procedure for finding the stage-wise critical decision boundaries is greatly simplified. The method also permits data-driven adaptations (using data from previous stages) to both the sample size and the statistical test, which offers new opportunities to speed up testing for evoked response detection.

Ontogeny of auditory brainstem responses in the bat, Phyllostomus discolor

Hearing is the primary sensory modality in bats, but its development is poorly studied. For newborns, hearing appears essential in maintaining contact with their mothers and to develop echolocation abilities. Here we measured auditory brainstem responses (ABRs) to clicks and narrowband tone pips covering a large frequency range (5-90 kHz) in juveniles (p7 to p200) and adults of the bat, Phyllostomus discolor. Tone-pip audiograms show that juveniles at p7 are already quite responsive, not only below 20 kHz but up to 90 kHz. Hearing sensitivity increases further until about p14 and is then refined, possibly correlated with growth and differentiation of the animals' outer ears. ABR amplitudes decrease within the first 3-4 months, inversely correlated with the bat weight and forearm length. ABR Wave I latency decreases with increasing stimulation level. ABR duration (measured between Waves I and V) is longer in juveniles and shortens with age which may reflect temporal refinement of auditory brainstem neurons to accommodate the exceptional temporal precision required for effective echolocation. Overall our data show that P. discolor bats have good hearing very early in life. The current method represents a fast and minimally invasive way of characterizing basic hearing in bats.

Assessing recurrent interactions in cortical networks: Modeling EEG response to transcranial magnetic stimulation

BACKGROUND

The basic mechanisms underlying the electroencephalograpy (EEG) response to transcranial magnetic stimulation (TMS) of the human cortex are not well understood.

NEW METHOD

A state-space modeling methodology is developed to gain insight into the network nature of the TMS/EEG response. Cortical activity is modeled using a multivariariate autoregressive model with exogenous stimulation parameters representing the effect of TMS. An observation equation models EEG measurement of cortical activity. An expectation-maximization algorithm is developed to estimate the model parameters.

RESULTS

The methodology is used to assess two different hypotheses for the mechanisms underlying TMS/EEG in wakefulness and sleep. The integrated model hypothesizes that recurrent interactions between cortical regions are the source of TMS/EEG, while the segregated model hypothesizes that the TMS/EEG results from excitation of independent cortical oscillators. The results show that the relatively simple EEG response to TMS recorded during non-rapid-eye-movement sleep is described equally well by either the integrated or segregated model. However, the integrated model fits the more complex TMS/EEG of wakefulness much better than the segregated model.

COMPARISON WITH EXISTING METHOD(S)

Existing methods are limited to small numbers of cortical regions of interest or do not represent the effect of TMS. Our results are consistent with previous studies contrasting the complexity of TMS/EEG in wakefulness and sleep.

CONCLUSION

The new method strongly suggests that effective feedback connections between cortical regions are required to produce the TMS/EEG in wakefulness.

Bootstrap Signal-to-Noise Confidence Intervals: An Objective Method for Subject Exclusion and Quality Control in ERP Studies

Analysis of event-related potential (ERP) data includes several steps to ensure that ERPs meet an appropriate level of signal quality. One such step, subject exclusion, rejects subject data if ERP waveforms fail to meet an appropriate level of signal quality. Subject exclusion is an important quality control step in the ERP analysis pipeline as it ensures that statistical inference is based only upon those subjects exhibiting clear evoked brain responses. This critical quality control step is most often performed simply through visual inspection of subject-level ERPs by investigators. Such an approach is qualitative, subjective, and susceptible to investigator bias, as there are no standards as to what constitutes an ERP of sufficient signal quality. Here, we describe a standardized and objective method for quantifying waveform quality in individual subjects and establishing criteria for subject exclusion. The approach uses bootstrap resampling of ERP waveforms (from a pool of all available trials) to compute a signal-to-noise ratio confidence interval (SNR-CI) for individual subject waveforms. The lower bound of this SNR-CI (SNR_LB) yields an effective and objective measure of signal quality as it ensures that ERP waveforms statistically exceed a desired signal-to-noise criterion. SNR_LB provides a quantifiable metric of individual subject ERP quality and eliminates the need for subjective evaluation of waveform quality by the investigator. We detail the SNR-CI methodology, establish the efficacy of employing this approach with Monte Carlo simulations, and demonstrate its utility in practice when applied to ERP datasets.

Consciousness and Complexity during Unresponsiveness Induced by Propofol, Xenon, and Ketamine

A common endpoint of general anesthetics is behavioral unresponsiveness, which is commonly associated with loss of consciousness. However, subjects can become disconnected from the environment while still having conscious experiences, as demonstrated by sleep states associated with dreaming. Among anesthetics, ketamine is remarkable in that it induces profound unresponsiveness, but subjects often report "ketamine dreams" upon emergence from anesthesia. Here, we aimed at assessing consciousness during anesthesia with propofol, xenon, and ketamine, independent of behavioral responsiveness. To do so, in 18 healthy volunteers, we measured the complexity of the cortical response to transcranial magnetic stimulation (TMS)--an approach that has proven helpful in assessing objectively the level of consciousness irrespective of sensory processing and motor responses. In addition, upon emergence from anesthesia, we collected reports about conscious experiences during unresponsiveness. Both frontal and parietal TMS elicited a low-amplitude electroencephalographic (EEG) slow wave corresponding to a local pattern of cortical activation with low complexity during propofol anesthesia, a high-amplitude EEG slow wave corresponding to a global, stereotypical pattern of cortical activation with low complexity during xenon anesthesia, and a wakefulness-like, complex spatiotemporal activation pattern during ketamine anesthesia. Crucially, participants reported no conscious experience after emergence from propofol and xenon anesthesia, whereas after ketamine they reported long, vivid dreams unrelated to the external environment. These results are relevant because they suggest that brain complexity may be sensitive to the presence of disconnected consciousness in subjects who are considered unconscious based on behavioral responses.

Recovery of cortical effective connectivity and recovery of consciousness in vegetative patients

Patients surviving severe brain injury may regain consciousness without recovering their ability to understand, move and communicate. Recently, electrophysiological and neuroimaging approaches, employing simple sensory stimulations or verbal commands, have proven useful in detecting higher order processing and, in some cases, in establishing some degree of communication in brain-injured subjects with severe impairment of motor function. To complement these approaches, it would be useful to develop methods to detect recovery of consciousness in ways that do not depend on the integrity of sensory pathways or on the subject's ability to comprehend or carry out instructions. As suggested by theoretical and experimental work, a key requirement for consciousness is that multiple, specialized cortical areas can engage in rapid causal interactions (effective connectivity). Here, we employ transcranial magnetic stimulation together with high-density electroencephalography to evaluate effective connectivity at the bedside of severely brain injured, non-communicating subjects. In patients in a vegetative state, who were open-eyed, behaviourally awake but unresponsive, transcranial magnetic stimulation triggered a simple, local response indicating a breakdown of effective connectivity, similar to the one previously observed in unconscious sleeping or anaesthetized subjects. In contrast, in minimally conscious patients, who showed fluctuating signs of non-reflexive behaviour, transcranial magnetic stimulation invariably triggered complex activations that sequentially involved distant cortical areas ipsi- and contralateral to the site of stimulation, similar to activations we recorded in locked-in, conscious patients. Longitudinal measurements performed in patients who gradually recovered consciousness revealed that this clear-cut change in effective connectivity could occur at an early stage, before reliable communication was established with the subject and before the spontaneous electroencephalogram showed significant modifications. Measurements of effective connectivity by means of transcranial magnetic stimulation combined with electroencephalography can be performed at the bedside while by-passing subcortical afferent and efferent pathways, and without requiring active participation of subjects or language comprehension; hence, they offer an effective way to detect and track recovery of consciousness in brain-injured patients who are unable to exchange information with the external environment.

Auditory evoked potentials for monitoring during anaesthesia: a study of data quality

The auditory evoked potential termed the middle latency response (MLR) has been suggested as an indicator of adequacy of anaesthesia during surgery. However, the response is small and must be extracted from high levels of background noise. A key consideration in using the MLR for clinical monitoring is whether data quality is sufficient to detect small changes. The aim of this study was to investigate the quality of the MLR recorded during anaesthesia, as a rigorous analysis of data quality is lacking in many studies. MLR recordings from patients sedated in intensive care after cardiac surgery were compared to recordings from a reference group of young volunteers with normal hearing. Data quality was measured with the F(sp) parameter. A bootstrap analysis was used to measure statistical response presence and to detect within-subject changes during clinical anaesthesia. Noise levels were high in the normative group probably due to myogenic and EEG activity. With 5 Hz click stimulation, MLR presence in the normative group was below 30%. Response presence improved using stimulation paradigms with chirps or maximum length sequences and reached 100% with a combination of maximum length sequences and chirps. F(sp) values generally improved during anaesthesia as noise levels reduced and MLR presence was 100% for MLS click stimulation. Changes in the MLR amplitude with propofol infusion rate were small. Some within-subject changes in MLR amplitude were detected using the bootstrap analysis, but 100% detection was not possible.

CONCLUSION

Obtaining good quality MLR data in awake subjects is challenging. Data quality improves during clinical anaesthesia and with advanced stimulation methods, but reliable detection of changes in the MLR for clinical monitoring remains a challenge.

'Weight of Evidence' analysis of neonatal sensory evoked potentials

Abnormalities in neonatal sensory evoked potentials (EPs) may indicate a poor developmental prognosis, but such EPs are highly variable, changing with development, and requiring subjective analysis. 'Weight of Evidence' (W), the logarithm of the ratio of the probability that a response has occurred to the probability that it has not, and 'Response Entropy' (S), the spread of the response over time and frequency bands, might provide objective and quantitative measures of EP abnormalities and developmental changes, based on information processing characteristics. W and S were calculated from visual and somatosensory EPs recorded in 72 premature newborns over 2 sessions, separated by 6-9 weeks. Group 1 had normal brain ultrasound images at the time of recording, and a normal developmental outcome at age 2 years. Group 2 had abnormal brain ultrasound images but normal outcome. Group 3 had abnormal brain imaging and abnormal outcome.W values were lowest in Group 3 (visual p<0.001: somatosensory p<0.04). Entropy diminished between sessions (visual p<0.001: somatosensory p<0.015): it was highest in Group 2 (visual p<0.03). The low W in Group 3 implies a lower signal/noise ratio, reducing information capacity. Decreasing entropy suggests more efficient information encoding with maturation.

Objective threshold estimation and measurement of the residual background noise in auditory evoked potentials of goldfish

A survey of papers using auditory evoked potentials (AEPs) published over the last 10 years (Table I) demonstrates that most AEP studies in animals have used subjective methods for auditory threshold determination. Subjective methods greatly reduce the value of statistical hypothesis testing and jeopardize tests of hypothetical experimental group differences in hearing sensitivity. Correspondingly, many attempts have been made to develop objective threshold determination methods, but these have not been used widely. Further, they seldom include an appreciation of the effects of residual noise in the AEP. In this study, AEPs evoked by tonal and noise stimuli in goldfish (Carassius auratus) were recorded and the residual background noise was measured and analyzed in detail. High variability was found in residual noise, but can be effectively controlled with a simple modification of averaging routines. Considerable interobserver disagreements were found using subjective threshold estimation. An objective method of threshold determination was developed based on comparison between AEP amplitude and controlled residual noise, using a signal detection theory approach to set specific threshold criteria. The usefulness of AEP in hypothesis testing for auditory function requires more control over residual background noise amplitudes and the use of objective threshold determination techniques.

An objective assessment of fetal and neonatal auditory evoked responses

OBJECTIVE

We propose to use cross-correlation function to determine significant fetal and neonatal evoked responses (ERs).

METHODS

We quantify ERs by cross-correlation between the stimulus time series and the recorded brain signals. The statistical significance of the correlation is calculated by surrogate analysis. For validation of our approach we investigated a model which mimics the generation of ERs. The model assumes a fixed latency of the ER and contains two parameters, epsilon and lambda. Whether or not the system responds to a given stimulus is controlled by epsilon. The amount to which the system is excited from the base line (background activity) is governed by lambda. We demonstrate the technique by applying it to auditory evoked responses from four fetuses (21 records) between 27 and 39 weeks of gestational age and four neonates (eight records).

RESULTS

The method correctly identified the ER and the latency incorporated in the model. A combined analysis of fetuses and neonates data resulted in a significant negative correlation between age and latency.

CONCLUSIONS

The analysis of ER, especially for fetal and newborn recordings, should be based on advanced data analysis including the assessment of the significance of responses. The negative correlation between age and latency indicates the neurological maturation.

SIGNIFICANCE

The proposed method can be used to objectively assess the ER in fetuses and neonates.

Modeling GABA alterations in schizophrenia: a link between impaired inhibition and altered gamma and beta range auditory entrainment

The disorganized symptoms of schizophrenia, including severely disordered thought patterns, may be indicative of a problem with the construction and maintenance of cell assemblies during sensory processing and attention. The gamma and beta frequency bands (15-70 Hz) are believed relevant to such processing. This paper addresses the results of an experimental examination of the cortical response of 12 schizophrenia patients and 12 control subjects when presented with auditory click-train stimuli in the gamma/beta frequency band during measurement using magnetoencephalography (MEG), as well as earlier work by Kwon et al. These data indicate that control subjects show an increased 40-Hz response to both 20- and 40-Hz stimulation as compared with patients, whereas schizophrenic subjects show a preference for 20-Hz response to the same driving frequencies. In this work, two computational models of the auditory cortex are constructed based on postmortem studies that indicate cortical interneurons in schizophrenic subjects have decreased GAT-1 (a GABA transporter) and GAD(67) (1 of 2 enzymes responsible for GABA synthesis). The models transition from control to schizophrenic frequency response when an extended inhibitory decay time is introduced; this change captures a possible effect of these GABA alterations. Modeling gamma/beta range auditory entrainment in schizophrenia provides insight into how biophysical mechanisms can impact cognitive function. In addition, the study of dynamics that underlie auditory entrainment in schizophrenia may contribute to the understanding of how gamma and beta rhythms impact cognition in general.

Bootstrap significance of low SNR evoked response

In order to obtain adequate signal to noise ratio (SNR), stimulus-evoked brain signals are averaged over a large number of trials. However, in certain applications, e.g. fetal magnetoencephalography (MEG), this approach fails due to underlying conditions (inherently small signals, non-stationary/poorly characterized signals, or limited number of trials). The resulting low SNR makes it difficult to reliably identify a response by visual examination of the averaged time course, even after pre-processing to attenuate interference. The purpose of this work was to devise an intuitive statistical significance test for low SNR situations, based on non-parametric bootstrap resampling. We compared a two-parameter measure of p-value and statistical power with a bootstrap equal means test and a traditional rank test using fetal MEG data collected with a light flash stimulus. We found that the two-parameter measure generally agreed with established measures, while p-value alone was overly optimistic. In an extension of our approach, we compared methods to estimate the background noise. A method based on surrogate averages resulted in the most robust estimate. In summary we have developed a flexible and intuitively satisfying bootstrap-based significance measure incorporating appropriate noise estimation.