In an effort to contribute to the improvement of Matric Biology education, a survey was conducted in 1996 and 1997 of Matric pupils and first year students at several tertiary institutions, in order to identify those topics which learners found most difficult. Photosynthesis and respiration were among the topics with which Matric Biology learners experienced many conceptual difficulties. The aim of this project was twofold: firstly to identify specific misconceptions students had regarding these topics, and secondly, to develop and evaluate a learning tool that would address these misconceptions. In order to identify the most common specific misconceptions, a quantitative research approach was taken. A three-tiered multiple choice questionnaire was developed, and administered to first year students in the 1998 intakes at MLSultan Technikon and the Biology Department of the University of Natal, Durban. It was also administered to students at the end of their first, second and third years of Cell Biology (University of Natal), Analysis of the questionnaires revealed that students did not understand the complementary relationship between photosynthesis and respiration. Computer-based virtual worlds provide constructivist learning environments, in which visualisation and problem solving in a complex system is possible. It was proposed that use of a virtual world would be an effective means of addressing the misconceptions we identified. A game was developed that presented students with authentic tasks of filling an oxygen cylinder (as an air supply) and a carbon dioxide cylinder (which would later be used to extinguish a fire). In order to do this students were required to solve a series of three puzzles, all of which related to the processes of photosynthesis and respiration. To account for different learning styles, the puzzles were based on three of Gardner's multiple forms of intelligence. Evaluation of the virtual world made use of a combination of quantitative and qualitative research methods. Students' understanding of the processes was measured with the use of the questionnaire. A deeper evaluation of their understanding and affective response to the game was obtained through interviews. It was found that students who had played the game had a clearer understanding of the complementary relationship between photosynthesis and respiration, and understood that respiration in plant cells is a continuous process. Students also showed greater confidence in their grasp of the processes, and reported that playing the game had been an enjoyable way of complementing their traditional lecture material in order to master these concepts. The virtual world was an effective learning tool for addressing the misconceptions students held regarding photosynthesis and respiration. / Thesis (M.Sc.)-University of Natal, Durban, 1998.

Le syndrome d’apnée obstructive du sommeil affecte 6% à 17% de la population adulte. Le traitement de référence est la ventilation nocturne par une pression positive continue (PPC) fixe ou autopilotée afin de maintenir les voies aériennes ouvertes. L’efficacité de traitement des PPC autopilotées dépend des algorithmes et technologies pour détecter et qualifier les événements respiratoiresDes bancs d’essai ont été créés pour évaluer les PPC autopilotées en conditions comparables, en simulant des scénarios respiratoires composés de chaînes répétitives d’événements respiratoires. Les profils respiratoires simulés par les bancs d’essai précédents sont standardisés et simplifiés par rapport au profil respiratoire du patient.Pour tendre vers des essais plus réalistes, un nouveau banc d’essai physiologique permettant de reproduire automatiquement les profiles respiratoires à partir des données polygraphiques a été créé pendant cette thèse. Il a été validé en évaluant la simulation de scénarios respiratoires de différents phénotypes issus de 12 patients.Via ce banc d’essai, la précision de l’index d’apnée-hypopnée (IAH) résiduel fourni par PPC a été évalué, en comparant les IAH déterminés par 4 dispositifs de PPC (AirSense 10, DreamStation Auto, S.Box et Prisma 20A) avec ceux de polygraphie. Les résultats ont permis de quantifier les différences d’IAH afin d’aider les médecin à en tenir compte. / Obstructive sleep apnea syndrome affects 6% to 17% of adult population. The reference treatment is nocturnal ventilation via an either fixed or auto-titrating positive airway pressure (APAP) to maintain upper airway (UA) open. Treatment efficiency of APAP depends greatly on algorithms and technologies used for detecting and characterizing disordered breathing events (linked to UA obstruction or central command).Bench tests have been developed to evaluate APAP devices under the same conditions, by simulating respiratory scenarios composed of a repetitive string of several disordered breathing events registered from apneic patients or artificially designed.Therefore, breathing profiles simulated on benches are standardized and simplified, in comparison with patients’. To improve this disadvantage, a new physiological bench, which enables reproducing automatically a specific patient breathing profile from its polygraph recordings, has been created. It has been validated by simulating various breathing profiles issuing from 12 patients of different pathological phenotypes.Through this new bench, the accuracy of residual apnea hypopnea indices (AHI) determined by 4 APAP devices (AirSense 10, DreamStation Auto, S.Box and Prisma 20A) has also been investigated in comparison with polygraph scorings. The results would help physicians in clinical practice thanks to the quantification of AHI discrepancies between manufacturers.