The impact of Central Auditory Processing Disorders (CAPD) on language and communication of children and adults is well documented in research. Several tests and normative data exist to assess and diagnose for CAPD, but these tests are not valid for use in South Africa due mismatch in the language or accent of the test recording and the language spoken in South Africa as well as normative data that were collected in the United States of America. As such CAPD may remain undiagnosed, misdiagnosed or designed management plans may be inappropriate as they arose from invalid tests. This study employed a developmental study design in order to adapt an existing 2-pair dichotic digit test (2-pair DDT) to a South African English accent to better suit the South African context. The study further investigated the effect of accent on dichotic digit test performance by comparing the performance of South African adults on the newly adapted South African accent 2-pair DDT and the existing 2-pair DDT, specified by the Department of Veteran Affairs (DVA), using a Wilcoxon-signed ranked test. Lastly, the study collected normative data for the 2-pair DDT that can be used in South Africa. The study successfully used a developmental study design to produce a South African accent 2-pair DDT that meets the international specifications as outlined by the DVA. A Wilcoxon signed ranked test showed that South Africans performed better on the South African accent 2- pair DDT than on the DVA specified test. The latter finding was supported by qualitative reports from the participants who found the DVA specified 2-pair DDT stimuli to be more difficult to listen to as compared to the South African accent test. Furthermore, the study produced preliminary normative data for the South African accent 2-pair DDT that met international standards and may be used to score 2-pair DDT performance. The successful use of the developmental study design to adapt the 2-pair DDT shows promising possibilities as a systematic process for adapting or developing tests that are more contextually appropriate for various contexts.

Much is known about the nature and causes of certain types of hearing loss. Although different etiologies of hearing loss may produce similar threshold impairments, as seen on the pure tone audiogram, their effects on the ability to discriminate speech and speakers under less than optimal listening situations are not well defined. It is well established that measures of speech intelligibility obtained in quiet do not provide an accurate assessment of this ability in everyday listening situations. Thus, the purpose of the present investigation was to assess the speech discrimination abilities of normal hearing and hearing impaired listeners in both quiet and noise, in order to study the effects of hearing loss, noise and speakers on speech understanding and speaker intelligibility. Initially, speech discrimination ability was assessed for elderly and young normal hearing individuals in quiet and in noise at comfortable listening levels, using a male and female speaker who were found to produce significant differences in intelligibility in noise in a previous study (Gengel & Kupperman, 1980) of young normal hearing listeners. A significant speaker interaction was found in noise between groups, where the elderly group obtained higher speech discrimination scores with the male speaker and the young, normal hearing group obtained higher scores with the female speaker. The overall performance of the elderly group was significantly poorer than that of the normal hearing group in both quiet and noise. Three additional groups of listeners (acoustic trauma, filtered normal hearing and unfiltered normal hearing) were then assessed and another (female) speaker and softer noise presentation level were included to further investigate speaker and noise interactions. The results obtained indicate that the overall performance of acoustic trauma, filtered normal hearing, and elderly listeners in noise is similar and that the performance of young normal hearing groups in noise is significantly different from these hearing impaired groups. Although differences in group performance scores were apparent, the greater variability of test scores obtained in noise limits the prediction of discrimination ability in noise on the basis of hearing loss. In addition, it was determined that the pure one audiogram did not predict discrimination performance by speaker in noise, although the relationship between performance in noise and high frequency hearing loss was confirmed. Significant speaker interactions were also noted in noise for the three additional subject groups tested. In general, consistent speaker differences were obtained for all groups in noise. The hearing impaired groups obtained higher discrimination scores in noise with the male speaker and the normal hearing groups obtained higher discrimination scores with one of the female speakers. However, the acoustic trauma group did not conform to this general trend of the hearing impaired listeners in the higher noise level condition. The results obtained support the need for the standardization of speaker and conditions for the clinical assessment of speech discrimination performance in noise. Three vocal parameters (intensity, fundamental frequency, and word duration) were assessed for each of the three speakers and correlated with the error scores extracted from 18 of the 50 words presented which produced the greatest dispersion of differences between speakers. Although significant differences were found between speakers for each of the three parameters measured, they were not found to be highly related to the differences in speech intelligibility produced. However, the interactive effects of a speaker's vocal pitch and a listener's hearing loss are postulated. The implications of this investigation and the directions of future research are discussed.

Includes bibliographical references (leaves 84-93).

Includes bibliographical references.

Bibliography: leaves 96-101.

Includes bibliographical references (leaves 196-214). / This research study stemmed from my experience as a community-based Audiologist in the Mitchell's Plain sub-district in Cape Town, South Africa. During this time, I worked closely with the local school nurses and witnessed first-hand the many challenges that these health workers are faced with. Through my collaborative work with the Mitchell's Plain school health nurses I identified the urgent need for a contextually relevant hearing screening protocol to assist the school nurses in improving the current state of the school-based hearing screening service. I therefore embarked on this research process taking on the role of a 'clinician-researcher' who had prior relationships with the primary research participants i.e. the local school nurses.

Includes abstract. Includes bibliographical references.

Includes bibliographical references (leaves 116-131).

Includes abstract. Includes bibliographical references.

Introduction: The computerized rotational head impulse test (crHIT) was recently developed to overcome limitations of the video head impulse test (vHIT) and assess the functioning of the lateral semicircular canals (SCC) in a more objective manner. Instead of an examiner applying rapid, brief, and unpredictable head rotations, as during the vHIT, the crHIT utilizes computer-controlled whole-body rotations. Since the crHIT is newly developed, its test-retest reliability for clinical use needed to be determined. Methods: Thirty-one healthy adult participants, between the ages of 18 and 40, with normal lateral SCC functioning and no symptoms or history of vestibular dysfunction were assessed with the crHIT using both earth-bound, stationary and head-fixed targets. These participants were assessed on three separate occasions: the second evaluation took place one to six hours after the initial evaluation and the third evaluation 24 hours to two weeks after the second evaluation. Results: For the stationary target, the mean angular vestibulo-ocular reflex (aVOR) gain ranged from 0.98 to 1.02 at a 95% confidence interval. A one-way repeated measures ANOVA test was performed to provide information about the reliability and repeatability of the aVOR gain across the different time intervals. It was demonstrated that the mean aVOR gain values were statistically equal across all three time intervals, and between both protocols that utilized stationary targets. Additionally, Bland-Altman plots demonstrated high Limits of Agreement (LoA). For the head-fixed target, a one-way repeated measures ANOVA demonstrated significant differences in the aVOR gain values both across and within the three testing sessions. These significant differences were demonstrated as a decrease in aVOR gains both across and within the three testing sessions. Conclusion: The crHIT is a reliable measure of determining lateral SCC function. Further investigation is required to determine the physiological mechanisms underlying the decreased aVOR gain values across multiple testing sessions. However, it is suspected that vestibular habituation as well as aVOR and saccadic adaptation mechanisms have potentially contributed to the decrease in aVOR gain values across the multiple testing sessions. / Dissertation (MA (Audiology))--University of Pretoria, 2021. / Speech-Language Pathology and Audiology / MA (Audiology) / Restricted