Steady-state auditory evoked potentials to amplitude-modulated tones in hearing-impaired subjects

Free-Field Cortical Steady-State Evoked Potentials in Cochlear Implant Users

Auditory steady-state evoked potentials (SS-EPs) are phase-locked neural responses to periodic stimuli, believed to reflect specific neural generators. As an objective measure, steady-state responses have been used in different clinical settings, including measuring hearing thresholds of normal and hearing-impaired subjects. Recent studies are in favor of recording these responses as a part of the cochlear implant (CI) device-fitting procedure. Considering these potential benefits, the goals of the present study were to assess the feasibility of recording free-field SS-EPs in CI users and to compare their characteristics between CI users and controls. By taking advantage of a recently developed dual-frequency tagging method, we attempted to record subcortical and cortical SS-EPs from adult CI users and controls and measured reliable subcortical and cortical SS-EPs in the control group. Independent component analysis (ICA) was used to remove CI stimulation artifacts, yet subcortical responses of several CIs were heavily contaminated by these artifacts. Consequently, only cortical SS-EPs were compared between groups, which were found to be larger in the controls. The lower cortical SS-EPs' amplitude in CI users might indicate a reduction in neural synchrony evoked by the modulation rate of the auditory input across different neural assemblies in the auditory pathway. The brain topographies of cortical auditory SS-EPs, the time course of cortical responses, and the reconstructed cortical maps were highly similar between groups, confirming their neural origin and possibility to obtain such responses also in CI recipients. As for subcortical SS-EPs, our results highlight a need for sophisticated denoising algorithms to pinpoint and remove artifactual components from the biological response.

Human auditory steady-state responses

Steady-state evoked potentials can be recorded from the human scalp in response to auditory stimuli presented at rates between 1 and 200 Hz or by periodic modulations of the amplitude and/or frequency of a continuous tone. Responses can be objectively detected using frequency-based analyses. In waking subjects, the responses are particularly prominent at rates near 40 Hz. Responses evoked by more rapidly presented stimuli are less affected by changes in arousal and can be evoked by multiple simultaneous stimuli without significant loss of amplitude. Response amplitude increases as the depth of modulation or the intensity increases. The phase delay of the response increases as the intensity or the carrier frequency decreases. Auditory steady-state responses are generated throughout the auditory nervous system, with cortical regions contributing more than brainstem generators to responses at lower modulation frequencies. These responses are useful for objectively evaluating auditory thresholds, assessing suprathreshold hearing, and monitoring the state of arousal during anesthesia.

Pure-tone threshold prediction by 80-Hz amplitude-modulation following response

The usefulness of 80-Hz amplitude-modulation following response (AMFR) detected by phase spectral analysis to predict the hearing threshold during sleep was evaluated in 20 normal adults, 8 normal children and 37 children with hearing impairment. The onset effect of tonal stimulus on 80-Hz steady state response was studied in normal adults during sleep and a threshold of 80-Hz AMFR detected by phase spectral analysis was compared with that of ABR elicited by tone pips in children during sleep. Although 80-Hz AMFR is not appropriate for the assessment of hearing in adults, it appears useful for evaluating hearing in young children during sleep. Hearing prediction by 80-Hz AMFR appears to be more accurate than that by ABR elicited with tone pips. The onset effect of stimulus on 80-Hz AMFR with modulation depth of 95% was less than 80-Hz SSR evoked by clicks.

Frequency specificity of amplitude-modulation-following response detected by phase spectral analysis

To investigate the frequency specificity of steady-state responses elicited with a sinusoidally amplitude-modulated tone (amplitude-modulation-following response, AMFR), AMFR at different carrier frequencies were examined in 15 subjects with normal hearing, 12 patients with abrupt high-frequency hearing loss, 5 patients with low-frequency hearing loss and 1 patient with mid-frequency hearing loss. The stimulus was a sinusoidally amplitude-modulated tone with a modulation frequency of 40 Hz and modulation depth of 95%. Carrier frequencies were from 250 to 4000 Hz in normal subjects and from 250 to 8,000 Hz in hearing-impaired subjects. Fridman's phase spectral analysis was conducted for automatic detection of response. The thresholds of AMFR were distributed within 30 dB nHL at all carrier frequencies in normal subjects. In the hearing-impaired subjects, AMFR thresholds were between 0 and 25 dB above the behavioral threshold at all carrier frequencies. The threshold patterns of AMFR detected by phase spectral analysis very closely resembled the corresponding audiogram patterns in all types of hearing impairment. AMFR is thus shown to have a very good frequency specificity.

Amplitude-modulation following response (AMFR): effects of modulation rate, carrier frequency, age, and state

Scalp responses to continuous amplitude-modulated (AM) tones were recorded from adults and 1-month-old infants. The amplitude-modulation following (or envelope) response (AMFR) was quantified using magnitude-squared coherence. This measurement indicates the strength of the frequency-following response relative to background neural noise. The optimal modulation rate for generating the AMFR was determined by studying the effects of stimulus modulation rate on the response. Stimulus AM rate was varied between 10 and 80 Hz for continuous tonal stimuli of 500 Hz, and between 20 and 80 Hz for continuous tonal stimuli of 2000 Hz. Optimal modulation rate was defined as the AM rate that provided the highest coherence estimate. Adult AMFR coherence increased between 10 and 40 Hz (20-40 Hz for 2000 Hz), and decreased between 40 and 80 Hz in both carrier frequency conditions. Infant AMFR coherence, in contrast, monotonically increased between 10 and 80 H (20-80 Hz for 2000 Hz). Thus, within the frequency range examined, 40 Hz is optimal for generating the AMFR in adults, whereas 80 Hz is optimal in infants. Adults were tested while awake and infants were tested during periods of sleep. Given the observed age difference in effective modulation rate, we examined modulation rate effects in a group of adults in both awake and sedated states. As in sleeping infants, 80 Hz was optimal for generating AMFRs in the sedated adults.

40-Hz steady-state responses in newborns and in children

The authors investigated the 40-Hz steady-state responses (SSR) in 32 full-term newborns and in 10 normal children (5-8 years old), using 500-Hz tone bursts. The 40-Hz SSR threshold is located at about 50 and 30 dB nHL in newborns and older children, respectively. The latencies of both P1 and N1 waves decreased significantly with age, while the amplitudes increased. No significant latency and amplitude intersex differences have been observed. Moreover, with age, the 40-Hz SSR became more stable, their test-retest replicability improved, and P1-N1 wave occurrence increased. The authors finally discuss the possible underlying mechanisms of these findings and conclude that the 40-Hz SSR are difficult to obtain and are scarcely reliable in defining the low-frequency threshold in newborns. The stability and reliability of the responses increase with age, and the electrophysiological and behavioral thresholds to low-frequency stimuli tend to overlap.