The corticofugal oscillatory modulation of the cochlear receptor during auditory and visual attention is preserved in tinnitus

Abstract

Introduction: The mechanisms underlying tinnitus perception are still under research. One of the proposed hypotheses involves an alteration in top-down processing of auditory activity. Low-frequency oscillations in the delta and theta bands have been recently described in brain and cochlear infrasonic signals during selective attention paradigms in normal hearing controls. Here, we propose that the top-down oscillatory activity observed in brain and cochlear signals during auditory and visual selective attention in normal subjects, is altered in tinnitus patients, reflecting an abnormal functioning of the corticofugal pathways that connect brain circuits with the cochlear receptor.

Methods: To test this hypothesis, we used a behavioral task that alternates between auditory and visual top-down attention while we simultaneously measured electroencephalogram (EEG) and distortion-product otoacoustic emissions (DPOAE) signals in 14 tinnitus and 14 control subjects.

Results: We found oscillatory activity in the delta and theta bands in cortical and cochlear channels in control and tinnitus patients. There were significant decreases in the DPOAE oscillatory amplitude during the visual attention period as compared to the auditory attention period in tinnitus and control groups. We did not find significant differences when using a between-subjects statistical approach comparing tinnitus and control groups. On the other hand, we found a significant cluster in the delta band in tinnitus when using within-group statistics to compare the difference between auditory and visual DPOAE oscillatory power.

Conclusion: These results confirm the presence of top-down infrasonic low-frequency cochlear oscillatory activity in the delta and theta bands in tinnitus patients, showing that the corticofugal suppression of cochlear oscillations during visual and auditory attention in tinnitus patients is preserved.

Keywords: EEG; attention; auditory efferent; corticofugal; oscillations; tinnitus.

FIGURE 1

(A) Behavioral task used to evaluate visual and auditory attention. Left: Visual attention: Subjects had to report the position of the clock’s tick at the offset of the green circle (visual attention cue). Simultaneously two tones (f1 and f2) that elicit DPOAEs were presented as auditory distractors. Right: Auditory attention: Individuals had to respond to the gap of silence embedded in the f1 and f2 stimuli while the clock‘s tick is moving randomly. DPOAE were recorded continuously during the whole task. (B) Circular histograms of behavioral responses during the visual attention task, in which 0° represents the clock’s tick target position. The magnitude of the column bars illustrates the probability of a given response position in the clock referenced to the visual target offset. The dashed lines mark the mean angle for each group (Tinnitus: 0.98°, CI-95%: [–1.10°, 3.07°]; Control: –5.03°, CI-95%: [–6.79°, –3.26°]).

FIGURE 2

Grand average hearing thresholds. Bilateral average audiogram thresholds from 0.125 kHz to 16 kHz in controls (blue, n = 14) and tinnitus (red, n = 14) subjects. Error bars represents SEM. Although high-frequency hearing threshold tended to be worse in tinnitus, this difference did not reach statistical significance in the present sample.

FIGURE 3

Box-plots showing the median latencies of responses to visual and auditory targets in the behavioral task. In the auditory task, the latencies for detecting the gap in DPOAE are shown. Meanwhile, in the visual task in the visual task, the latencies correspond to the report of the visual cue (green circle) offset.

FIGURE 4

Grand average time spectra calculated for control and tinnitus groups during visual and auditory attention tasks. The color intensity scale was normalized to z-scores in all conditions. The vertical line in the panels illustrate time “0” in auditory and visual tasks. Note the presence of low-frequency oscillatory activity in theta and delta bands (< 10 Hz) in EEG and DPOAE channels during the selective attention period (from 0 to 1,500 ms). A significant reduction in the magnitude of the oscillatory activity in the cochlear channel (DPOAE) was observed in control and tinnitus groups during the visual attention period as compared to the auditory attention period. An increase in the theta power is observed in the Fz channel during the visual attention period in tinnitus.

FIGURE 5

Time spectra average differences between auditory and visual periods using permutation tests in a within subjects approach. Color represents the grand average of the differences in Z-score. Red and blue clusters represent significant differences in the permutation test, while non-significant differences are shown in green. The first and third column show uncorrected results, while the second and fourth illustrate multiple comparison corrected results. Positive significant differences are represented in red (i.e., auditory > visual), while negative differences are illustrated in blue (i.e., auditory < visual). Note that after correction, there is a significant positive cluster in the delta-theta band in the DPOAE channel of the tinnitus group, and a significant negative cluster in the theta band in the Cz channel of the control group.

FIGURE 6

Time spectra average differences between auditory and visual periods using permutation tests in a between subjects approach. Color represents the grand average of the differences in Z-score. The first and third columns show uncorrected results, while the second and fourth columns illustrate multiple comparison corrected results. The uncorrected results reveal differences in the delta and theta bands in EEG channels, whereas the multiple comparison corrected analyses yielded non-significant differences (non-significant differences are depicted in green).

FIGURE 7

Temporal course of the amplitude of 1–8 Hz frequency band between EEG and DPOAE channels and comparison of slope values between individuals (*p < 0.05).