Cochlear implant users perceive mainly sound amplitude modulation cues. Processing of these amplitude modulations can be subject to interferences, so that the perception of a modulation in a target sound can be impaired by a superimposed sound if this sound contains a similar modulation. Such phenomenon, which is observed both in subjects with normal-hearing and in cochlear-implant users, could be explained by difficulties to direct attention to relevant information in complex sound signals. Selective auditory attention also plays a crucial role in speech comprehension in cocktail-party situations where the speech of multiple talkers get mixed at the ear of a listener. Cochlear implant users typically struggle in these cocktail-party situations and report intense listening effort.

The present clinical trial aims at evaluating the contribution of selective auditory attention for sound modulations to the listening effort of patients with cochlear implants and of healthy volunteers with normal-hearing during speech perception under cocktail-party-like conditions.

Selective auditory attention abilities of patients and controls will be assessed using a psychoacoustical test whereby their ability to detect a target sound amplitude modulation will be measured both in the absence and in the presence of an interfering (i.e. distracting) amplitude modulation occurring in a distant spectral region from that of the target. The effect of this distractor's presence on modulation detection performance will serve as a behavioural index of the subject's auditory attention capacities.

The attentional capacity index will then be tested as a predicting factor for the listening effort of the subject during a speech-in-noise consonant identification task. Listening effort will be measured from the pupil dilation response to the presented speech units (pseudowords).

This study will enhance our understanding of cochlear implant user's perception and listening effort and will serve as a basis for prognostic tests of listening effort and of implantation success for cochlear implant candidates, based on a simple measurement of auditory attentional abilities.

Participants to this trial will be adult patients with cochlear implants (experimental group) and adults with normal-for-the-age hearing (healthy volunteers or controls), recruited in a single center (Reference Centre for Cochlear Implants in Adults, Otorhinolaryngology Service, Auditory Implants Functional Unit, Pitié-Salpêtrière Hospital Group). All patients followed-up at the clinical center and satisfying the inclusion criteria will be addressed a mail informing them about the study. They will be contacted by phone after a reflexion delay of 15 days to check their understanding of the study and probe their agreement or refusal to participate. Healthy controls will be recruited by a call broadcasted through mailing lists dedicated to volunteers for scientific experiments. Those who answer the call by phone or email will be convoked by order of contact, according to the study's needs, and given information about the study. Signed informed consent will be collected from all participants prior to participation, on the day of inclusion, as well as their demographic data, educational level and, for patients with cochlear implants, the date of implantation and hearing loss etiology. Each subject will be identified by a unique series of 9 alphanumerical characters. Healthy volunteers will undergo tonal audiometry and will be definitively included in the study only after normal hearing for the age is confirmed.

The entire participation is planned to take place in a single session lasting 1 hour and 40 minutes, on the day of inclusion, including consent collection, preliminary measurements (CT thresholds for patients, audiometry for controls), material calibration, psychoacoustic tests and speech recognition tests with oculometry. Only as an exception in cases where the entire session could not be completed in the allocated time, participants will be invited to come back for a second session on a different day to finish incomplete tests.

All tests will be performed in a soundproofed booth. Participants will sit in front of a screen displaying a visual fixation point during trials, as well as relevant information (end of trial, response alternatives, performance feedback). Pupil diameter will be recorded during the tests using an eye-tracker device (Tobii TX300). Prior to the tests, ambient luminosity will be adjusted so that the subject's pupil be at the center of its measured dynamics. Cochlear implant patients will be equipped with an implant processor dedicated to the experiment (reference processor), ensuring their own processor's settings won't be modified and that the tests be performed with a well-functioning processor.

The psychoacoustic tests will consist in an amplitude modulation detection task with a 3-interval, 3-alternative forced choice design. In each trial, participants will be presented with 3 successive sounds in a random order, only one of which constrains the target amplitude modulation and which the participant must report. All presented sounds will contain two superimposed sinusoidal carriers (e.g. 500 Hz and 4000 Hz). The target amplitude modulation will be applied to only one these two carriers (the target carrier), in only one of the three presented sound.

The settings of the reference processor during these tests will deactivate all electrodes instead of two, making sure that the stimulation is maximally focused in two, well separated regions of the cochlea. The acoustic frequency of these active electrodes might be modified compared to the usual processor, so as to guarantee that they will each be maximally activated by one of the two stimulus carriers. The dynamics of the active electrodes will be set with the own C and T thresholds of the patient, and sound calibration will ensure that the average stimulation level correspond to 50% of the CT dynamics. Stimuli will be presented to patients through the analogic input of the reference processor and through headphones, on a single, randomly-drawn side (left or right), at an average level of 70 decibel Sound Pressure Level, to the healthy volunteers.

The test will comprise two conditions presented alternately. In the first condition, there will be no other modulation than the target. In the second condition, a distracting modulation will be applied to the non-target carrier, in the three intervals. Both target and distracting modulations will have the same 8-Hz rate. Performance in these two conditions will be assessed the measurement of amplitude modulation detection thresholds (AMDT) using an adaptive procedure, repeated twice per condition. The difference in performance between the two conditions will serve as an index of the auditory attentional capacity of the subject.

Participants will perform a training followed by the AMDT measurements, as well as tests with constant modulation levels and joint pupil diameter measurement, all of which will be distributed in blocks meant to last between 5 and 8 minutes and interleaved by breaks for a total of about 40 minutes.

The final, speech recognition tests will consist in a consonant identification task. The settings of the processor in these tests will be identical to the patient's processor usual settings, and sounds will be presented in free field through loudspeakers, at a comfortable level (65 decibel Sound Pressure Level for target speech and 55 decibel Sound Pressure Level for interfering speech). In each trial, participants will be presented with the recording of a pseudoword, pronounced by a female voice, of the form [aCaCa] where [C] is a consonant to identify among 16 alternatives. This test will be performed once in silence and a second time in the presence of interfering pseudospeech pronounced by a different voice, with a signal-to-noise ratio of +10 decibel. There will be 48 trials in each condition. Pupil diameter will be recorded during the test so as to measure listening effort in each condition as the mean pupil dilation response with reference to a pre-stimulus baseline. The difference of this pupil dilation response between the two conditions will serve as a measurement of the amount of listening effort induced by the presence of interfering speech.

During data analysis, the linear correlation between the auditory attention capacity index (from the psychoacoustic test) and the interference-induced listening effort measurement (from the speech recognition test) will be tested by means of a Pearson test, separately for each of the two subject groups (cochlear implant patients and normal-hearing healthy volunteers), with a 0.05 p-value significance threshold.

Inclusion Criteria:

Arm 1: patient with Cochlear Implant

• Having a unilateral Oticon Medical cochlear implant
• Duration of use of the implant of 6 months or more
• Disyllabic word recognition score of at least 40% on average in silence, to limit "floor" effects in hearing tests.
• Disyllabic word recognition score not exceeding 30% on the contra-lateral ear alone, aided or not, in silence.

Arm 2: volunteers with normal hearing

- Normal tonal audiometry for the age

For both

• Age between 18 and 80 years old
• Mother tongue : French
• Normal vision with or without correction
• Absence of eye pathologies (cataracts, nystagmus, amblyopia, macular degeneration).
• Not taking psychotropic drugs or drugs affecting the parasympathetic nervous system
• Absence of pathology or neurological history (especially head trauma, stroke).
• Information and signing of a consent prior to any act related to research

Exclusion Criteria:

For both groups, ensuring the reliability of the pupillometric measurement (Winn et al., 2018):

• Pathologies of the eye: cataracts, nystagmus, amblyopia, macular degeneration.
• Taking psychotropic drugs and / or drugs affecting the parasympathetic nervous system
• Neurological pathology (in particular head trauma or stroke) associated with an alteration of cognitive functions or history thereof that may affect the stability of the gaze, the congruence of eye movements, pupillary dilation or with an alteration of cognitive functions

Others criteria:

• No affiliation (or being entitled) to a social security scheme
• Person under State Medical Assistance
• Person under legal protection (tutorship, curatorship, other…) or under family authorization