Peripheral deficits and phase-locking declines in aging adults

Abstract

Age-related difficulties in speech understanding may arise from a decrease in the neural representation of speech sounds. A loss of outer hair cells or decrease in auditory nerve fibers may lead to a loss of temporal precision that can affect speech clarity. This study's purpose was to evaluate the peripheral contributors to phase-locking strength, a measure of temporal precision, in recordings to a sustained vowel in 30 younger and 30 older listeners with normal to near normal audiometric thresholds. Thresholds were obtained for pure tones and distortion-product otoacoustic emissions (DPOAEs). Auditory brainstem responses (ABRs) were recorded in quiet and in three levels of continuous white noise (+30, +20, and +10 dB SNR). Absolute amplitudes and latencies of Wave I in quiet and of Wave V across presentation conditions, in addition to the slope of Wave V amplitude and latency changes in noise, were calculated from these recordings. Frequency-following responses (FFRs) were recorded to synthesized /ba/ syllables of two durations, 170 and 260 ms, to determine whether age-related phase-locking deficits are more pronounced for stimuli that are sustained for longer durations. Phase locking was calculated for the early and late regions of the steady-state vowel for both syllables. Group differences were found for nearly every measure except for the slopes of Wave V latency and amplitude changes in noise. We found that outer hair cell function (DPOAEs) contributed to the variance in phase locking. However, the ABR and FFR differences were present after covarying for DPOAEs, suggesting the existence of temporal processing deficits in older listeners that are somewhat independent of outer hair cell function.

Keywords: Auditory aging; Auditory brainstem response; Frequency-following response; Peripheral deficit; Phase locking.

Fig. 1.

Audiometric and distortion product otoacoustic emissions (DPOAEs) thresholds are elevated in older normal-hearing listeners (ONH, red triangles) compared to younger normal-hearing listeners (YNH, blue circles) across the frequency range, but the group differences widen above 3 kHz (audiogram) and above 2. 6 kHz (DPOAEs). Error bars: ± 1 Standard Error.

Fig. 2.

Auditory nerve function. The average auditory brainstem response (ABR) waveforms in the left-most plot (obtained with derived horizontal montage) show that overall amplitude is lower in ONH (red) compared to YNH (blue) listeners. Shaded regions: ± 1 Standard Error. The notched box plots compare Wave I amplitude, Wave V/I ratio, and Wave I latency between ONH and YNH participants. Wave I amplitude is significantly higher in the ONH compared to the YNH participants. **p < 0.01.

Fig. 3.

Aging effects on Wave V. The average ABR waveforms (vertical montage) obtained in quiet and in white noise at +30, +20, and +10 dB SNR show that in ONH listeners, overall amplitudes are lower and Wave V latencies are delayed across conditions compared to the YNH listeners. Shaded regions: ± 1 Standard Error. The dashed line in each plot was placed at the mean latency in the YNH listeners.

Fig. 4.

Noise effects on Wave V. Top panel: Average ABR waveforms across quiet and decreasing SNR conditions are overlaid separately in ONH and OHI groups. Changes in latency and amplitude are apparent in the YNH listeners but not in the ONH listeners. Bottom panel: Notched box plots are displayed for the slope of change in Wave V with decreasing SNRs for latency and amplitude in YNH and ONH listeners. The ONH listeners have shallower slopes than the YNH listeners for the amplitude decrease, but the slopes are not statistically different between the groups for the latency increase.

Fig. 5.

Top panel: Stimulus waveforms are displayed for the shorter 170-ms /ba/ and the longer 260-ms /ba/. Bottom panel: Average response waveforms corresponding to the shorter and longer /ba/ stimuli are displayed for the YNH (blue) and ONH (red) listeners. Note that the periodicity of the stimuli is mirrored in the responses. An age-related reduction in response amplitude is apparent in the response waveforms.

Fig. 6.

Top panel: Average phase-locking factor (PLF) to the temporal envelope of the shorter 170-ms /ba/ and the longer 260-ms /ba/ represented in the time-frequency domain, with hotter colors representing higher phase locking in YNH and ONH listeners. Age-related PLF reductions are observed for both syllabi. Bottom panel: Notched box plots are displayed for the PLF corresponding to the early and late regions of the steady-state vowels of the /ba/ stimuli in YNH (blue) and OHI (red) listeners.

Fig. 7.

Scatter plots demonstrating relationships among phase-locking factor (PLF) and distortion-product otoacoustic emission average (DPAVG) and Wave I amplitude across groups and within male (blue) and female (red) groups. DPAVG was positively correlated with PLF. *p < 0.05. Shaded region: Confidence interval (α=0.05).