Selective attention to visual stimuli reduces cochlear sensitivity in chinchillas

Abstract

It is generally accepted that during periods of attention to specific stimuli there are changes in the neural activity of central auditory structures; however, it is controversial whether attention can modulate auditory responses at the cochlear level. Several studies performed in animals as well as in humans have attempted to find a modulation of cochlear responses during visual attention with contradictory results. Here, we have appraised cochlear sensitivity in behaving chinchillas by measuring, with a chronically implanted round-window electrode, sound-evoked auditory-nerve compound action potentials and cochlear microphonics, a measure of outer hair cell function, during selective attention to visual stimuli. Chinchillas were trained in a visual discrimination or in an auditory frequency discrimination two-choice task. We found a significant decrease of cochlear sensitivity during the period of attention to visual stimuli in the animals performing the visual discrimination task, but not in those performing the auditory task, demonstrating that this physiological effect is related to selective attention to visual stimuli rather than to an increment in arousal level. Furthermore, the magnitude of the cochlear-sensitivity reductions increased in sessions performed with shorter target-light durations (4-0.5 s), suggesting that this effect is stronger for higher attentional demands of the task. These results demonstrate that afferent auditory activity is modulated by selective attention as early as at sensory transduction, possibly through activation of olivocochlear efferent fibers.

Figure 1.

Visual discrimination task. A, Schematic diagram of the front panel of the operant cage. B, Time sequence of the task. I–III, A neutral cue (central light) was turned on before one of two lateral lights (targets); the chinchilla had five seconds to respond by pressing a lever under the target light. IV, Simultaneously an irrelevant click and/or tone train was presented at 1–4 Hz rate. V, The round-window electrical signal was acquired during the task. Inset, Modified task in which only 70% of the trials began with the neutral cue.

Figure 2.

CAP results. A, Example of CAP reductions in correct-response trials (solid squares) compared with omission trials (open circles) during the period of visual attention (in this and the following figures; light gray, neutral-cue period; dark-gray, target-light period). Symbols on each trace represent CAP amplitude changes measured in decibels, referenced to the average amplitude of the potentials measured before the onset of the neutral cue. We show CAPs from a single recording session (100 trials) in response to a 2 Hz click train. Significant CAP reductions, calculated for pairs of values measured from the neutral-cue onset up to the mean response latency, are indicated by asterisks (unpaired t test; p < 0.05). Accuracy was 95% and mean response latency, always measured from the target-light onset, was 1688 ms (vertical segmented line). B, Progressively greater CAP reductions for gradual decrease of target-light duration (4–0.5 s) in one of two tested animals showing correlation between increase in attentional demands of the task and amount of CAP reduction. The curves shown in this panel for the different target durations correspond to the first recording session of 100 trials (of two daily sessions). The gray bar (arrow) indicates the interval (1274–2109 ms) of mean correct-response latencies measured from the target-light onset in the 4 d. C, Correlation between target duration and CAP reduction. We show data from two chinchillas (solid and open symbols) in which we successfully reduced the target duration from 4 s to 0.5 s during successive recording sessions. We found a significant correlation between target duration and CAP reduction (p < 0.01, Spearman test). The equation of the linear regression shown in this figure is y (dB) = −0.5 × (s) + 2.75; (R² = 0.73). In B and C, only sessions in which chinchillas had accuracies >75% and made at least 20 correct responses are shown.

Figure 3.

Results for compound stimuli. A, Example of CM augmentation (solid squares) concomitant to CAP reductions (open circles) during the period of visual attention. Symbols on each trace represent CAP and CM amplitude changes measured in decibels, referenced to the average amplitude of the respective cochlear potential measured before the onset of the neutral cue. The figure also shows a histogram for the latencies of correct responses to the target (dark-gray shaded area; scale in right y-axes). The top right inset shows examples of averaged CAP and CM traces from −500 ms (thin trace) and 2000 ms (thick trace; see vertical arrows in A; calibration: 0.5 ms, 40 μV for CAP, 80 μV for CM). In this case, accuracy was 80.2% and mean response latency measured from the target-light onset was 1833 ms (vertical segmented line). B, Control of sound pressure level within the chinchilla's bulla using a microphone. Changes of sound pressure (triangles), CAP decreases (open circles), and CM augmentations (solid squares) during the period of visual attention (symbols on each trace represent changes measured in decibels, as in A). Notice that neither CAP decreases nor CM increases are caused by changes in sound pressure level reaching the cochlea. Accuracy was 85.3% and mean response latency measured from the target-light onset was 2034 ms (vertical segmented line).

Figure 4.

Reliability of the effect in cochlear potentials. A, B, Examples of CAP reductions (A) and CM augmentations (B) recorded in one chinchilla during 5 consecutive days in identical stimulus conditions. In all cases, compound stimuli were delivered at 2 Hz rate and accuracy was >75%. The gray bars with arrows indicate the interval (1907–2091 ms) of mean correct-response latencies measured from the target-light onset in the 5 d. Symbols on each trace represent cochlear potentials amplitude changes measured in decibels, referenced to the average amplitude of the respective cochlear potential measured before the onset of the neutral cue.

Figure 5.

Results for the modified 70/30 visual task. A, B, CAP reductions (A) and CM augmentations (B) in the modified task (solid squares, correct responses in the 70% of trials with the neutral cue; open circles, correct responses in the 30% without the neutral cue; triangles, incorrect responses in the 70% of the trials with the neutral cue; solid diamonds, omission trials). Significant CAP reductions and CM increases in correct trials compared with omission trials are indicated by asterisks and by crosses for trials with and without the neutral cue, respectively (unpaired t test, p < 0.05). Notice that incorrect trials with the neutral cue also present CAP reductions and CM increases similar to those in the correct trials with the neutral cue. Mean correct-response latencies measured from the target-light onset, in trials with and without the neutral cue are shown with segmented lines (1520 ms, cued trials; 2832 ms, no cue trials). The mean incorrect-response latency was 1103 ms.

Figure 6.

Auditory frequency discrimination two-choice task. A, Timeline of the auditory task. I, II, In 70% of the trials (I), the neutral cue was presented before the onset of a target tone, whereas (II) in the other 30% of the trials the target tone was presented without the neutral cue. III, Auditory stimuli. The compound stimuli (cs; a click of alternating polarity followed after 40 ms by a 1200 Hz tone) were presented at a rate of 1 Hz. The reference tone (2000 Hz, 190 ms) was presented before one of the target tones (tone for pressing left lever, 1200 Hz; right lever, 3600 Hz; duration, 3810 ms). IV, Chinchillas had a response window of 300–5000 ms from the onset of the reference tone. V, Simultaneously, we recorded the round window signal during behavior. B, Mean response latency to the target tone measured from the reference tone onset, with and without neutral cue. The mean response latencies were shorter for the 70% of the trials in which the neutral cue was presented (with neutral cue, 1619 ms; without neutral cue, 2409 ms; unpaired t test, p < 0.01). C, CAP and CM amplitude changes (in decibels) for correct-trial responses, referenced to the mean values measured previous to the onset of the neutral cue. There were no changes in CM and CAP amplitudes during the presentation of the neutral cue when the targets were tones.