Thesis (Ph. D.)--University of Texas at Austin, 2001. / Vita. Includes bibliographical references (leaves 137-139). Available also in a digital version from Dissertation Abstracts.

Recent work has shown that preschoolers track informants’ past reliability concerning familiar information and labels, and they use this information to judge the correctness of novel information and labels they provide. But linguistic factors also sway children’s choices for social interaction, for which native-accented speakers are preferred. The present study uses the selective trust paradigm to consider how accentedness interacts with speaker reliability with native- and foreign-accented informants. The results show that speaker reliability and accentedness affect four-year-olds’ choices, but the impact of these factors differed by response type. Preschoolers preferred to ask the native-accented speaker for information, regardless of his reliability. However, in choosing which label to learn, preschoolers selected the reliable speaker’s label, regardless of accent, and correctly identified the unreliable speaker. This study provides evidence suggesting that young children separate their social biases from their objective assessment of novel information. / text

Not available / text

This dissertation addresses the modeling of wind turbine systems (WTS) and wind farms. The WTS reliability model provides the generation state space of a WTS. The generation states are derived from the combinations of the wind states from given wind data and the condition states of each component in a WTS. Wake effect is accounted when there are neighboring WTSs. The results of the reliability model of a WTS are associated with the generation states of the WTS, which include the probability, transition rates to other states, frequency of transitions to other states, and duration. The reliability model of the wind farm is derived by combining the wind states, WTS states and the distribution line states. The results of the reliability model of a wind farm are associated with the generation states of the wind farm, which include the probability, transition rates to other states, frequency to other states, and duration. The reliability model of the wind turbine system and the reliability model of the wind farm presented in this dissertation bring contribution to the planning and operation of bulk power systems with wind farm integration. The developed models can provide the system operator with clear reliability indices in terms of generation states of wind turbine systems and wind farms along with their probability, duration and frequency of transitions.

This thesis is an investigation into the use of simulation for the study of quality economics within a cell-based manufacturing environment. Quality is, for the purposes of this report, considered to mean conformance to specification. This research identified two distinct areas of study that could be considered part of quality economics. The economic design of quality control procedures is concerned with the "micro" side of quality economics, focusing on single processes, whilst quality costing systems represent the "macro" side, focusing on entire production lines, departments or organisations. A critical analysis of these areas suggested a number of deficiencies. In addition, a number of economic quality models were considered. These were found to focus on the "macro" side of quality economics, with no consideration of the effect of specific changes to specific processes - the "micro" side. It was suggested that simulation would be a suitable approach for overcoming some of these deficiencies. The precedent for the use of simulation to investigate quality economics is considered, with the conclusion that none of the existing approaches reviewed were able to model the manufacturing systems under consideration in sufficient detail. Therefore, a new approach is outlined, and the thesis then describes the development and validation of this approach. A number of examples are given that are intended to both illustrate the simulation approach and, hopefully, provide certain generic information that can be applied to any similar manufacturing process. Conclusions are drawn regarding the success of the project relative to the initial aims, and the contribution to our understanding of quality economics. Finally, a number of recommendations for further work are given.

University-built satellite programs are prone to failure because these projects are performed by inexperienced student-engineers during the early parts of the satellite-building “learning curve”. However, with sufficient attention on risk management, students should be able to identify what risk avoidance actions should be taken, and when. By applying risk mitigation strategies, university built satellite programs will not only contribute students to learn space systems engineering, but also accomplish their scientific missions with higher rates of success. This thesis study is aimed to provide risk management guidelines that could be adapted to university built satellite programs to increase the risk awareness. Besides indicating the key strategies for risk mitigation, a set of risk management procedures are prepared to help students during the university-built satellite projects. / text

Future computing platforms will increasingly demand more stringent memory resiliency mechanisms due to shrinking memory cell size, reduced error margins, higher capacity, and higher reliability expectations. Traditional mechanisms, which apply error checking and correcting (ECC) codes uniformly across all memory locations, are inefficient -- Uniform protection dedicates resources to redundant information and demand higher cost for stronger protection, a fixed (worst-case based) error tolerance level, and a fixed access granularity. The design of modern computing platforms is a multi-objective optimization, balancing performance, reliability, and many other parameters within a constrained power budget. If resiliency mechanisms consume too many resources, we lose an opportunity to improve performance. Hence, it is important and necessary to enable more efficient and flexible memory resiliency mechanisms. This dissertation develops techniques that enable efficient, adaptive, and dynamically tunable memory resiliency mechanisms. First, we develop two-tiered protection, apply it to the last-level cache, and present Memory Mapped ECC (MME) and ECC FIFO. Two-tiered protection provides low-cost error detection or light-weight correction in the common case read operations, while the uncommon case error correction overhead is off-loaded to main memory namespace. MME and ECC FIFO use different schemes for managing redundant information in main memory. Both achieve 15-25% reduction in area and 9-18% reduction in power consumption of the last-level cache, while performance is degraded by only 0.7% on average. Then, we apply two-tiered protection to main memory and augment the virtual memory interface to dynamically adapt error tolerance levels according to user, system, and environmental needs. This mechanism, Virtualized ECC (V-ECC), improves system energy efficiency by 12% and degrades performance only by 1-2% for chipkill-correct level protection. V-ECC also supports ECC in a system with no dedicated storage for redundant information. Lastly, we propose the adaptive granularity memory system (AGMS) that allows different access granularities, while supporting ECC. By not wasting off-chip bandwidth for transferring unnecessary data, AGMS achieves higher throughput (by 44%) and power efficiency (by 46%) in a 4-core CMP system. Furthermore, AGMS will provide further gains in future systems, where off-chip bandwidth will be comparatively scarce. / text

No description available.

Photoreceptors in the eye basically provide information about light intensities from which brains extract different kinds of visual cues (e.g. color, movement, pattern). How do the properties and response characteristic of visual interneurons differ from the periphery to the central brain? I intracellularly recorded from neurons in the second and third optic ganglia (medulla and lobula) and the central brain (protocerebrum) of bees (mainly bumblebees; Bombus impatiens) while presenting color and motion stimuli. Bees rely on such stimuli during flight and foraging and show sophisticated visual learning abilities. We found that neurons in the distal medulla are color specific while ones in the proximal medulla show complex, often antagonistic color responses. Neurons in the distal lobula (layers 1-4) mainly process motion information while the proximal lobula (layers 5 and 6) seems to combine color and motion responses. Anterior parts of the central brain receive complex input representing combinations of motion and color information characterized by specific temporal properties (e.g. temporal precision, 'novelty' information or entrainment). This kind of often sparsely coded information is also represented in the mushroom bodies, learning and memory centers in the insect brain. In contrast, posterior parts of the central brain receive mainly motion information and show more reliable responses yet less precise spike timing. While the former kind of information (temporally precise or novelty in anterior pathways) is suited to form stimulus associations relevant during foraging, the latter, more reliable information is thought to support fast optomotor flight control maneuvers and other less plastic behaviors.

Frontal encephalographic (EEG) alpha asymmetry has been proposed as a measure of the relative difference in average cortical activity between the right and left anterior cortex, where this difference is taken as a physiological marker of trait and state level variables associated with affect. The validity of asymmetry as an indicator of both physiological and psychological variables is in part determined by the psychometric properties of asymmetry scores. The present studies focus on the psychometric assessment of frontal alpha asymmetry measured during rest. The first study involves a novel approach in the assessment of the internal consistency reliability of asymmetry scores. Previous studies estimated internal consistency reliability via Cronbach's alpha, using a relatively small set of asymmetry score that summarized activity over segments of the EEG data (e.g. one minute). Such an approach, however, will create estimates dependent on the number of segments utilized rather than the total amount of data recorded. Thus in the first study, individual FFT epochs were treated as items, thereby maximizing the total number of items used to estimate internal consistency reliability. Results of this study suggest internal consistency reliability is greater than previously reported, and as such, the duration of resting EEG data necessary to achieve a reasonable reliability criterion may be shorter than the current standard. In the second study, asymmetry scores were assessed as a specific case of difference scores, which are susceptible to a statistical artifact associated with differences in true-score variance for the component measures. Predicted asymmetry scores associated with the statistical artifact were obtained by estimating the true-score variance of right and left alpha power. The use of hierarchical linear regression showed some influence of the statistical artifact on the relationship between asymmetry scores and a measure of depressive severity, suggesting that some caution may be warranted in interpreting asymmetry results with relatively small effect sizes.