The primary contributions of this thesis are associated with the development of a new method for exploring the relationships between inputs and outputs for large scale computer simulations. Primarily, the proposed design space exploration procedure uses a hierarchical partitioning method to help mitigate the curse of dimensionality often associated with the analysis of large scale systems. Closely coupled with the use of a partitioning approach, is the problem of how to partition the system. This thesis also introduces and discusses a quantitative method developed to aid the user in finding a set of good partitions for creating partitioned metamodels of large scale systems. The new hierarchically partitioned metamodeling scheme, the lumped parameter model (LPM), was developed to address two primary limitations to the current partitioning methods for large scale metamodeling. First the LPM was formulated to negate the need to rely on variable redundancies between partitions to account for potentially important interactions. By using a hierarchical structure, the LPM addresses the impact of neglected, direct interactions by indirectly accounting for these interactions via the interactions that occur between the lumped parameters in intermediate to top-level mappings. Secondly, the LPM was developed to allow for hierarchical modeling of black-box analyses that do not have available intermediaries with which to partition the system around. The second contribution of this thesis is a graph-based partitioning method for large scale, black-box systems. The graph-based partitioning method combines the graph and sparse matrix decomposition methods used by the electrical engineering community with the results of a screening test to create a quantitative method for partitioning large scale, black-box systems. An ANOVA analysis of the results of a screening test can be used to determine the sparse nature of the large scale system. With this information known, the sparse matrix and graph theoretic partitioning schemes can then be used to create potential sets of partitions to use with the lumped parameter model.

This dissertation offers a new understanding about wartime decision making in the face of likely, but unintended, harm to foreign civilians. It empirically identifies conditions under which leaders in democratic nations are more or less likely to choose to attack a target when confronted with a dilemma between pursuing national security objectives and avoiding civilian casualties. An innovative targeting decision model was constructed that described both the theorized structure of the decisions inputs and the process by which these inputs are assembled into a choice. The model went beyond the normal target benefit and civilian casualty cost considerations of proportionality to also include the contextual input of prospect frame. Decision makers were expected to address the same benefit and cost differently depending on whether they were winning or losing the conflict. This was because the prospect frame would influence their risk attitudes, as predicted by prospect theory. This model was then tested via two decision-making experiments that used military officers and defense civilians as participants. Additionally, a statistical analysis of data collected from an extended period of the second Intifada was done to seek evidence that the model also applied in actual wartime decision making. All three tests supported portions of the targeting decision model. Higher target benefit and lower civilian casualty estimates increased support for the planned attack. Prospect frame influenced decisions in the cases where both target value and the civilian casualty estimates were high and the resulting dilemma was very difficult. In these situations, those told that their forces were losing the conflict were less sensitive to humanitarian harm and more likely to support the attack than when they were told their side was winning. Furthermore, the Intifada data analysis of attacks approved by Israeli officials against Palestinians found this same effect of prospect frame held generally across all six years of observations.

Disputats. Universität Hohenheim, 2002. / Haves kun i elektronisk udg. Thesis (Ph.D.). 41 ill., 5 tables; approx. 200 ref. Summaries (En, Ge).

Animal welfare is a topic subjected to great controversy, mostly within moral philosophy. The moral issue of human behaviour is often dealt with, alongside whether nonhuman animals are eligible certain rights. In our world, how humans behave towards nonhuman animals have fallen into something of political oblivion, which is the departure-point for this research. The essay’s discourse surrounds nonhuman animals’ political and moral status in Sweden, with the aim of drawing conclusions regarding whether they can be said to possess it. This is done through an analysis of ideas, using dimensions as tools, problematizing the Swedish parliamentary parties' views on animal welfare. A better understanding for nonhuman animals’ situation in Sweden has been provided, showing that there is no animal rights mentality tangible, while speciesist and utilitarian attitudes towards nonhuman animals dominate. The analysis show that nonhuman animals in Sweden possess moral status, as the parties agree that nonhuman animals should be spared from unnecessary suffering, and their welfare seems to count in its own right. However, human interests tend to take precedence in most cases, and in the end, nonhuman animals cannot be considered to possess any political status, despite their unmistakable presence in our society.

This dissertation is composed of three studies that use statistical, behavioral, and neuroimaging methods to investigate Chinese and English speakers’ semantic interpretations of four types of visual information including icons, single Chinese characters, single English words, and pictures. The goal is to examine whether people cognitively process icons as logographical words. By collecting survey data from 211 participants, the first study investigated how differently these four types of visual information can express specific meanings without ambiguity on a quantitative scale. In the second study, 78 subjects participated in a behavioral experiment that measured how fast people could correctly interpret the meaning of these four types of visual information in order to estimate the differences in reaction times needed to process these stimuli. The third study employed functional magnetic resonance imaging (fMRI) with 20 participants selected from the second study to identify brain regions that were needed to process these four types of visual information in order to determine if the same or different neural networks were required to process these stimuli. Findings suggest that 1) similar to pictures, icons are statistically more ambiguous than English words and Chinese characters to convey the immediate semantics of objects and concepts; 2) English words and Chinese characters are more effective and efficient than icons and pictures to convey the immediate semantics of objects and concepts in terms of people’s behavioral responses, and 3) according to the neuroimaging data, icons and pictures require more resources of the brain than texts, and the pattern of neural correlates under the condition of reading icons is different from the condition of reading Chinese characters. In conclusion, icons are not cognitively processed as logographical words like Chinese characters although they both stimulate the semantic system in the brain that is needed for language processing. Chinese characters and English words are more evolved and advanced symbols that are less ambiguous, more efficient and easier for a literate brain to understand, whereas graphical representations of objects and concepts such as icons and pictures do not always provide immediate and unambiguous access to meanings and are prone to various interpretations. / text

Paediatric clinical trials are crucial to ensure safety and efficacy of medicines in children. This study explored parents’ perceptions in Gaborone, Botswana, regarding children’s participation in clinical trials. One hundred participants completed a self-administered questionnaire. The results showed that 93% of the participants had a good knowledge of clinical trials; 74% thought that children would benefit from clinical trial participation; 63% would not enrol their children in clinical trials; 55% suggested that children should only be enrolled once they have reached the age of 18 years; and 71% reported that only children with an active disease should be enrolled. A large proportion (82%) reported that children’s participation should be voluntary, while joint parental consent was supported by 93% of responders. Regarding children’s assent 91% deemed respect for children’s assent essential, although 52% thought that children’s assent should not override the parental decision. There was a statistically significant correlation between finding clinical trials in general important and children’s participation in clinical trials (p=0.008, Fisher’s Exact Test), as well as the need for individual consent p<0.0001, Chi-Square). There was also a statistically significantly association between respondents, who would allow their children’s participation in clinical trials and who would encourage their family members and friends’ participation in clinical trials (p=0.0001, Chi-Square). An overwhelming 94% advocated for special regulations in Botswana to govern paediatric clinical trials. Almost all participants (99%) explicitly expressed the opinion that there should be global regulations for paediatric clinical trials. / Thesis (M.Soc.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2012.

This thesis seeks to demonstrate, by way of a multidisciplinary study, that consent is, despite its legal definition which refers to the free and enlighted expression of individual will, in fact, at times limited if not eliminated, by social considerations, arising from the medical, economic and legal context. These considerations reflect what one might call the social norm. An appropriate understanding of consent serves, therefore, to express the social norm as a constraint, which, in turn, acts as a measure of what it means to belong in society. Thus, while consent is often presented as the fundamental principle to be respected in biomedical research, it is in reality, merely one principle to respect among others. These limitations connected to consent are exacerbated in emergency situations where consent is sometimes reduced to mere signature, and in some cases it has been recognized that research can be undertaken without the subject's prior consent.

Le domaine des énergies est au cœur des préoccupations technologiques, politiques et économiques de notre société moderne. Ce domaine nécessite une compréhension minimale du concept scientifique de l’énergie. Elle est selon nous essentielle à toute formation citoyenne. Nous avons dans un premier temps, à partir de considérations théoriques et pratiques, examiné pourquoi ce domaine si important dans notre société technologique est si peu abordé dans le cursus scolaire québécois? Pourquoi se contente-t-on d’un enseignement théorique et discursif de ce concept? Pourquoi, au contraire de tout enseignement scientifique, n’a-t-on pas envisagé de situations d’apprentissages en laboratoire pour l’étude des énergies? Dans un deuxième temps, nous avons proposé une idée de solution concrète et réaliste pour répondre à l’ensemble de ces questions. Une solution qui invite les élèves à s’investir de manière constructive dans des activités de laboratoire afin de s’approprier ces concepts. Pour ce faire, nous avons conçu des variables globales énergies qui ont permis aux élèves de les mesurer et d’expérimenter facilement des transformations énergétiques. Cette recherche de développement technologique en éducation consiste donc à profiter des nouveaux développements technologiques de l’informatique et de la micro-électronique pour concevoir, réaliser et mettre à l’essai un environnement informatisé d’apprentissage en laboratoire pour les sciences et la technologie. Par ce que l’énergie est au confluent de trois domaines, cet environnement a été conçu pour supporter dans une même activité l’apprentissage des mathématiques, des sciences et de la technologie. Cette intégration recommandée par les nouveaux programmes est, selon nous, essentielle à la compréhension des concepts liés à l’énergie et à ses transformations. Par cette activité d’apprentissage multidisciplinaire, nous voulons, via une approche empirique et concrète, aborder ces problèmes de transformations énergétiques afin de donner aux élèves la capacité de perfectionner les prototypes qu’ils construisent en technologie de manière à améliorer leurs performances. Nous avons montré que cette démarche technoscientifique, assimilable à la conception d’un schème expérimental en sciences, favorise la compréhension des concepts liés aux énergies et à leurs transformations. Ce développement, ouvert à l’investigation scientifique, apporte un bénéfice didactique, non seulement, pour des enseignants en exercices et des étudiants-maîtres, mais aussi pour des élèves de 5ème année du niveau secondaire, ce que nous avons démontré dans une mise à l’essai empirique. / The energy sector is at the heart of the concerns technological, political and economic modern society. This area requires a basic understanding of the scientific concept of energy. It is our opinion essential to any citizen training. We initially, from theoretical and practical considerations, examined why this area is so important in our technological society is so little discussed in the Quebec curriculum? Why do we merely a theoretical and discursive concept? Why, contrary to all scientific education, have we not considered learning situations in the laboratory for the study of energy? In a second step, we proposed an idea of practical and realistic solution to address all these questions. A solution that invites students to engage constructively in laboratory activities to appropriate these concepts. To do this, we have developed global variables energies that allowed students to experiment and measure energy transformations easily. This quest for technological development in education is therefore to take advantage of new technological developments in computing and microelectronics to design, build and test a computerized environment learning laboratory for science and technology. With this energy is at the confluence of three domains, this environment has been designed to support the same activity in the learning of mathematics, science and technology. This integration recommended by the new programs we believe is essential to the understanding of concepts related to energy and its transformations. For this learning activity multidisciplinary, we want, via an empirical and practical address these issues of energy transformations in order to give students the ability to develop the prototypes they build technology to improve their performance. We have shown that this approach techno, similar to the design of an experimental design in science, promotes understanding of concepts related to energy and their transformations. This educational development, open to scientific investigation, provides educational benefit, not only for practicing teachers and student teachers, but also for students from the fifth year of high school, we have shown in a setting tested empirically.

High Performance Computing is preparing the era of the transition from Petascale to Exascale. Distributed computing systems are already facing new scalability problems due to the increasing number of computing resources to manage. It is now necessary to study in depth these systems and comprehend their behaviors, strengths and weaknesses to better build the next generation. The complexity of managing users applications on the resources conducted to the analysis of the workload the platform has to support, this to provide them an efficient service. The need for workload comprehension has led to the collection of traces from production systems and to the proposal of a standard workload format. These contributions enabled the study of numerous of these traces. This also led to the construction of several models, based on the statistical analysis of the different workloads from the collection. Until recently, existing workload traces did not enabled researchers to study the consumption of resources by the jobs in a temporal way. This is now changing with the need for characterization of jobs consumption patterns. In the first part of this thesis we propose a study of existing workload traces. Then we contribute with an observation of cluster workloads with the consideration of the jobs resource consumptions over time. This highlights specific and unattended patterns in the usage of resources from users. Finally, we propose an extension of the former standard workload format that enables to add such temporal consumptions without loosing the benefit of the existing works. Experimental approaches based on workload models have also served the goal of distributed systems evaluation. Existing models describe the average behavior of observed systems. However, although the study of average behaviors is essential for the understanding of distributed systems, the study of critical cases and particular scenarios is also necessary. This study would give a more complete view and under- standing of the performance of resource and job management. In the second part of this thesis we propose an experimental method for performance evaluation of distributed systems based on the replay of production workload trace extracts. These extracts, replaced in their original context, enable to experiment the change of configuration of the system in an online workload and observe the different configurations results. Our technical contribution in this experimental approach is twofold. We propose a first tool to construct the environment in which the experi- mentation will take place, then we propose a second set of tools that automatize the experiment setup and that replay the trace extract within its original context. Finally, these contributions conducted together, enable to gain a better knowledge of HPC platforms. As future works, the approach proposed in this thesis will serve as a basis to further study larger infrastructures.

Introduction Willingness to participate in clinical trials is a crucial element in recruitment of suitable participants for intervention trials. Measurement of willingness to participate assists in determining community preparedness for clinical trials, such as HIV vaccine trials. Therefore, researchers have developed a Clinical Research Involvement Scale (CRIS) to assess willingness to participate modelled on the Theory of Reasoned Action. The CRIS was tested in the USA and was noted that it would benefit from additional testing in other populations. Aim The purpose of this study is to determine whether racial differences exist in willingness to participate and explore potential factors associated with willingness to participate in HIV prevention research. Methods A cross sectional analytic study was conducted. The CRIS was administered to university students aged 18-45 at the University of KwaZulu-Natal in South Africa. The CRIS was administered online with a demographic questionnaire to facilitate evaluation of possible associations between willingness to participate and age, gender, relationship status, parity, religion, education status, student status, employment status and access to private health care. Participation was once-off at the time of completing the scale. Results The study enrolled 636 participants, two thirds being female. An effective sample size of 509 was considered for analysis after data was cleaned for accuracy and completeness. The results indicated that all students across all race groups were willing to participate in HIV prevention research. However, when considering factors that affected willingness to participate, statistically significant differences were noted. Based on the differences amongst these factors, Black students expressed greater intention to participate compared to White and Indian students. The CRIS was deemed a reliable instrument in this population; however in its current structure it did not show strong validity. Validity improved if the factors of motivation to comply and outcome evaluations were removed in this population. Discussion The study findings are specific to students of the University of KwaZulu-Natal and cannot be generalized to other populations. The racial differences in factors that affect willingness to participate indicate differences in risk perception and seeking access to better quality healthcare. Recommendations The CRIS should be used in other student populations to assess its validity. (350 words) / Thesis (M.Med.Sc.)-University of KwaZulu-Natal, Durban, 2013.