Tympanometry and acoustic stapedial reflex (ASR) are routinely used in audiology clinics to assess the functional integrity of the eardrum and middle ear system in humans. Conventional tympanometry (which delivers a probe tone of 226 Hz into the ear canal and measures the mobility of the eardrum as the air pressure in the ear canal is varied) and acoustic reflex testing are effective in detecting middle ear pathologies in children and adults. However, the clinical application of these two tests to infants younger than 7 months has major limitations. In recent years, high frequency tympanometry (HFT) with a probe tone of 1000 Hz has been trialled successfully in young infants (< 7 months) and research on ASRs as they apply to this age group is continuing. Although preliminary HFT data for this population are emerging, there has been no detailed study that describes the effect of age on HFT and ASR results, no clear guideline on ways to interpret the HFT results, and no investigation to measure the feasibility and reliability of the ASR findings. For these reasons, systematic investigation into the use of HFT and ASR measures for evaluating the middle ear function of young infants is warranted. This thesis aimed to: (i) investigate the feasibility of obtaining HFT and ASR findings from newborn and 6-week-old infants, and study the characteristics of the immittance findings in these two age groups; (ii) investigate methods within HFT to measure the middle ear admittance of newborn babies; (iii) establish normative HFT data from healthy newborn babies using the new component compensation method; (iv) examine the test-retest reliability of the ASR test in healthy neonates; and (v) investigate the test-retest reliability of the ASR test in 6-week-old infants. The aims of the thesis were met through five studies. In study one (Chapter 2), a pilot study was conducted to examine the feasibility of performing HFT and ASR in 42 healthy infants and study the characteristics of the immittance findings obtained from these infants using a longitudinal study design. In this pilot study, all infants were tested at birth and then re-tested approximately 6 weeks after the first test. This study confirmed the feasibility of obtaining valid immittance findings from healthy young infants. Most importantly, the findings of this pilot study revealed that the mean values of the majority of HFT parameters and acoustic stapedial reflex threshold (ASRT) obtained at 6 weeks were significantly greater than those obtained at birth, indicating the need to have separate sets of normative data for both tests for newborn and 6-week-old infants. In study 2 (Chapter 3), three different methods to measure middle ear admittance (often described as peak compensated static admittance) in 36 healthy neonates were compared. The three methods were the traditional baseline compensation method (compensated for the susceptance component at 200 daPa pressure) and two new component compensated methods (compensated for both the susceptance and conductance components at 200 daPa and -400 daPa). The results showed that the mean middle ear admittances obtained by compensating for the two components of admittance at a pressure of 200 daPa (YCC200) and -400 daPa (YCC-400) were significantly greater than that using the traditional baseline compensation method (YBC). The higher mean admittance results obtained using the new component compensated methods suggests that the two new methods have the potential to better separate normal from abnormal admittance results. The test-retest reliability of YBC, YCC200 and YCC-400 was investigated, with the result that a lower test-retest reliability was obtained for YCC-400 than for the other two measures. It was, therefore, concluded that the component compensation method compensated at 200 daPa may serve as an alternative method for estimating middle ear admittance, especially in the context of assessing neonates using HFT. In study 3 (Chapter 4), normative data were gathered using the new component compensation method (compensated at 200 daPa) on a group of 157 healthy newborn babies. In addition to the component compensated static admittance (YCC), normative data showing the 90 % ranges for tympanometric peak pressure, admittance at 200 daPa, uncompensated peak admittance, and traditional baseline compensated static admittance (YBC) were established in this study. No gender effect was found on any of the tympanometric measures. In study 4 (Chapter 5), the use of ASR to evaluate middle ear function in neonates was studied. The feasibility of obtaining ipsilateral ASR from neonates by stimulating their ears with a 2 kHz tone and broadband noise (BBN) was demonstrated. ASRs were elicited from 91.3% of 219 full-term normal neonates, while the remaining 8.7% of neonates who had flat tympanograms and no transient evoked otoacoustic emissions did not exhibit ASRs. Good test-retest reliability was demonstrated in the ASRT obtained using both the 2 kHz and BBN stimulus; there was no significant difference between test and retest conditions and intra-correlation coefficients of 0.83 for the 2 kHz tone and 0.76 for the BBN stimulus. In the last study (Chapter 6), the test-retest reliability of ASRT obtained from 70 6-week-old infants was investigated. The methodology described in Chapter 5 was followed. No significant difference in ASRT between test and retest conditions was found for the 2 kHz tone (mean ASRT = 67.3 dB HL versus 67.1 dB HL) and BBN stimulus (mean ASRT = 80.9 dB HL versus 81.6 dB HL). Good test-retest reliability of ASRT with intra-correlation coefficients of 0.78 was found for both the 2 kHz tone and the BBN stimulus. In essence, through achieving the aforementioned aims, the current research program was able to enhance the minimal literature available concerning the use of HFT and ASR testing in young infants. Ultimately, the findings presented in this thesis will inform clinicians of the recent developments in HFT and ASR testing, and assist them in evaluating the middle ear function of young infants with accuracy and confidence.

L’imagerie par résonance magnétique (IRM) est l’une des techniques les plus utilisées en neurosciences. Malheureusement, elle implique une exposition à des niveaux de bruit importants. Cette exposition peut entraîner des effets temporaires ou permanents sur l’audition, surtout lorsqu’elle est prolongée ou répétée. Plusieurs inconnus subsistent au sujet des risques pour l’audition de sujets subissant des examens en IRM à répétition. L’objectif de ce projet était de suivre six participants de recherche subissant des examens IRM hebdomadaires et d’évaluer sur une base régulière l’évolution de leur audition. L’audition des participants a été testée périodiquement sur une période d’un an et demi. Des valeurs de référence ont été récoltées pour chacun des tests avant le début de ce projet. Des séances de tests auditifs ont ensuite été planifiées immédiatement avant et après un scan ainsi qu’à la suite d’un délai variant entre deux et sept jours après un scan. Un article scientifique présentant les résultats les plus importants de ce projet a été rédigé. Il comporte les résultats à deux des tests administrés aux participants: l’audiométrie par tons purs et les émissions otoacoustiques par produit de distorsion. Aucun effet de l’exposition au bruit du scanneur n’a été révélé par ces tests. / Magnetic resonance imaging (MRI) is one of the most commonly used tools in neuroscience. However, it implies exposure to high noise levels. It is well known that exposure to noise can lead to temporary or permanent hearing damages, especially when the exposure is long or repeated. Little is known about the hearing risks of people undergoing several MRI examinations, especially in the context of prolonged longitudinal studies. The goal of this research project was to follow six research participants undergoing weekly MRI scan sessions and test the evolution of their auditory health on a regular basis. The participants’ hearing was tested periodically, over a period of a year-and-a-half. First, baseline values were acquired for each of the hearing tests before the beginning of this research project. Hearing tests were then scheduled immediately before/after a scan and with a delay between two and seven days after a scan. A scientific publication was written outlying the main results regarding two of the administered tests: pure-tone audiometry and distortion product otoacoustic emissions. No consistent impacts of the scanner noise exposure were found for those two tests.