The research reported in this thesis examines the relative contributions of top-down and bottom-up information during lexical access. I evaluate the Cohort Model of lexical access (Marslen-Wilson and Welsh, 1978; Marslen-Wilson and Tyler, 1980; and Marslen-Wilson, 1987) which specifies that the first stage in lexical access is fully autonomous and that during this stage all processing proceeds in terms of analysis of the acoustic-phonetic input. Implicit in this model is the assumption that bottom-up processing is immune to any effects of contextual or top-down information. I examine the extent to which listeners ever rely exclusively on bottom-up information during lexical access and investigate this issue empirically, by measuring effects of context on both the production and the perception of words in various contexts. I test the hypothesis that a word uttered in a constraining context will be acoustically indistinguishable from its competitors by, first, measuring one acoustic parameter (VOT) across constraining and non-constraining contexts and, then, examining the intelligibility of tokens of that parameter taken from the varyingly constraining contexts. The data from these experiments suggest that the realization of VOT is not an aspect of bottom-up information which would create problems for a bottom-up processor in terms of providing ambiguous acoustic-phonetic information. I then investigate whether bottom-up processing during lexical access is immune to effects of context. Following Grosjean (1980) and Tyler (1984), I utilize the Gating Paradigm. Using incongruous contexts, I argue that direct assessment of the contributions made by different information sources during lexical access can be made. By presenting bottom-up information which is inappropriate to the contextual (topdown) information, I evaluate the extent to which one information source is given priority over the other. I vary both the contextual constraints available to the listener and the acoustic clarity of bottom-up information. The observed pattern of listeners' identifications of the words suggested that whilst bottomup information was given priority, top-down information was available and was utilized during lexical access. I present data which support the working structure of the Cohort Model of lexical access. I conclude, however, that the model places disproportionate emphasis on initial bottom-up processing. It appears that top-down information is not prohibited from contributing to processing during the initial stage of lexical access.

Syftet med föreliggande arbete är att undersöka hur lärare i ett urval skolor talar om sitt arbete med den tidiga läs- och skrivundervisningen i årskurs F-3. Undersökningens fokus kommer i huvudsak vara lärares beskrivning av undervisningens utformning samt vilka faktorer och modeller som ligger till grund för deras läs- och skrivundervisning. Utifrån syftet görs en enkätundersökning där lärare yrkesverksamma i årskurs F-3 deltar. Undervisningsmodellerna Bornholmsmodellen, LTG-metoden, Att skriva sig till läsning, Wittingmetoden och Whole language ligger till grund för en del av enkätens frågor. Undervisningsmodellerna presenteras under arbetets forskningsbakgrund och enkätsvaren redovisas under arbetets empiriska del. Avslutningsvis leds arbetet in i en diskussion med utgångspunkt i arbetets syfte, forskningsfrågor och enkätsvar. Undersökningen visar att lärare använder sig av olika modeller i den tidiga läs- och skrivundervisningen. De slutsatser som lyfts fram i diskussionen är betydelsen av att kombinera olika undervisningsmodeller för att tillgodose elevers individuella behov och förutsättningar. Det visar sig även i arbetets undersökning att lärares yrkeserfarenhet är en faktor till hur lärarna väljer att utforma den tidiga läs- och skrivundervisningen.

A language comprehension mechanism that immediately starts processing language as it is encountered is typically thought of as one that speeds and facilitates spoken language comprehension. However, there exist cases where the earliest parts of a word or phrase encode information that is somewhat at odds with the remainder of the word or phrase in full. Examples of these "potentially misleading" cases include compound words where the initial subpart of the compound belongs to a different syntactic category than the entire compound (e.g., "popcorn", "greyhound"), or noun phrases where the initial element of the phrase signals perceptual properties possessed by the referent of the noun phrase (e.g., some Chinese Cantonese classifier-noun phrases). Using a visual-world methodology, this dissertation explores the kinds of unintended or "impulsive" activation that are triggered when listeners encounter such cases, as well as how syntactic and contextual cues can constraining this impulsive activation. Experiment 1 examines whether hearing compound subparts (e.g., "pop-" in "popcorn") activates conceptual associates across syntactic categories, and Experiment 2 examines whether this activation is moderated by listeners' expectations about the syntactic structure of the sentences they encounter. Experiments 3 and 4 investigate the processing of compounds whose initial subparts correspond to colour terms (e.g., "greyhound"). Experiment 3 explores whether these colour subparts trigger the activation of phrasal-level descriptions of non-target objects in the visual display, and whether this activation is influenced by the presence/absence of motivation to use colour descriptions when naming screen objects. Experiment 4 further explores whether a perceptual mismatch between a target object and the colour term in its name increases this impulsive activation. Finally, Experiment 5 investigates whether listeners use the perceptual information encoded in pre-nominal classifiers in Cantonese Chinese to guide their consideration of referential candidates whose perceptual properties do or do not match the classifier semantics. The findings from this dissertation point to the highly interactive nature of spoken language comprehension, suggesting that the kinds of impulsive activation under current discussion are rampant and automatic, but can also be suppressed to varying degrees by the syntactic, semantic, and contextual cues available to the listener.

Zugl.: Erlangen, Nürnberg, Univ., Diss., 2009

By visual attention process biological and machine vision systems are able to select the most relevant regions from a scene. The relevancy process is achieved either by top-down factors, driven by task, or bottom-up factors, the visual saliency, which distinguish a scene region that are different from its surrounding. During the past 20 years numerous research efforts have aimed to model bottom-up visual saliency with many successful applications in computer vision and robotics.In this thesis we have performed a comparison between a state-of-the-art saliency model and subjective test (human eye tracking) using different evaluation methods over three generated dataset of synthetic patterns and natural images. Our results showed that the objective model is partially valid and highly center-biased.By using empirical data obtained from subjective experiments we propose a special function, the Probability of Characteristic Radially Dependency Function, to model the lateral distribution of visual attention process.

Les savanes sont des écosystèmes complexes pilotées par plusieurs mécanismes ascendant (ex: les nutriments du sol ou pluviométrie) ou descendant (ex: feu ou herbivorie), mais l'importance relative de ces mécanismes reste largement débattue. En particulier, le rôle des herbivores brouteurs (browsers) reste mal compris en tant que source de perturbation, et donc de force de pression descendante influant sur la dynamique des savanes. Dans cette étude, deux approches ont été développées pour aborder le rôle des perturbations dans la dynamique des savanes. Dans un première partie, j'ai utilisé une approche comparative inter-site pour explorer les réponses des communautés de plantes, et des principaux traits de ces plantes associés aux feuilles, branches, architecture et défense, aux variations de quatre facteurs : les nutriments dans le sol, la pluviométrie, la pression d'herbivorie et l'intensité du feu. Seize sites de savane, en Afrique du Sud et au Zimbabwe, ont été sélectionnés sur des gradients de chacun de ces facteurs. Les espèces ligneuses dominantes (>80 % de la biomasse) sur chaque site ont été identifiées et échantillonnées, afin de mesurer les traits des feuilles et des branches associés à l'appétence, architecture, ainsi qu'aux défenses physiques et chimiques de ces plantes. Des mesures ont également été faites pour estimer les effets des meso-brouteurs et mega-brouteurs. Des transects ont permis d'estimer la fréquence et l'intensité du feu sur chaque site, et l'effet sur les plantes. En préambule à l'analyse, et devant le manque de protocole standard pour estimer la fertilité des sols dans la littérature écologique, je propose une méthode et un échantillonnage afin de définir de manière robuste la fertilité des sols sur chaque site. Dans cette partie inter-site, huit traits principaux ont été comparés sur le gradient de qualité de sol et de pluviométrie, et bien que quelques relations statistiques existent entre les traits des feuilles, le sol et la pluviométrie, ces relations sont très faibles comparées à celle trouvées dans les méta analyses inter-biomes publiées dans la littérature. Cependant, ces approches interbiome sont dominées par des sites tempérés qui ont des niveaux de perturbations bien inférieurs à ceux des savanes africaines. L'évaluation des effets des meso-brouteurs et mega-brouteurs le long des gradients de sol et de pluviométrie sur vingt traits associés aux défenses structurelles et chimiques des plantes montre que les défenses structurelles sont plus corrélées aux caractéristiques du sol que les défenses chimiques, mais que seules les défenses structurelles sont fortement corrélées à l'impact par les brouteurs. Le niveau d'utilisation des plantes par les mesobrouteurs apparaît plus prévisible en fonction des traits des plantes que celui par les mégabrouteurs. Dans une deuxième partie présentent des résultats de deux études basées sur des expériences d'exclos. Dans le parc national de Kruger, la composition de la communauté, l'abondance et la démographie des ligneux dominants ont été estimées à l'intérieur et à l'extérieur de trois exclos de 40 ans, et les brouteurs apparaissent comme ayant un impact significatif sur la distribution, la densité et la structure des populations des espèces arbustives et arborées ayant des traits préférés : forte concentration en azote foliaire et faible teneur en défenses chimiques. L'interaction entre les effets des brouteurs et du feu semble aussi affecter le recrutement des juvéniles ligneux dans les grandes classes de taille. Dans le parc de Hluhluwe-iMfolozi, cinq exclos ont été utilisé pour tester l'effet des brouteurs sur l'architecture, la croissance, les défenses chimiques et structurelles des jeunes individus de sept espèces d'acacia. Des différences nettes apparaissent entre les espèces d'acacia de savane semi-aride et plus humide dans les traits associés à l'appétence, l'architecture et les défenses … [etc] / Savannas are complex ecosystems affected by several bottom-up (e.g. soil nutrient availability and rainfall) and top-down (e.g. fire and herbivory) drivers. However, the relative importance of bottom-up vs. top-down drivers in influencing savanna dynamics is still widely debated. Within the top-down (disturbance) category of drivers, the role of mammal browsers in particular in driving savanna functioning is still not well understood. Two approaches were adopted to determine the role of disturbance in savannas. Firstly, by using a comparative approach, I attempted to address the so-called ‘savanna problem’ by investigating how savanna woody plant community compositions and key plant traits relating to the leaves, stems, architecture, and defence are influenced by soil nutrient status, rainfall, fire and browsing. Sixteen sites were selected along gradients of these four drivers from savanna parks throughout South Africa and Zimbabwe. The dominant woody species (species that accounted for >80% of standing biomass) at each site were identified and sampled for the key leaf and stem traits relating to plant functioning, palatability, architecture, physical and chemical defences. Measurements were undertaken for each species in order to determine both meso-browser and mega browser impact. Transects were undertaken in order to determine the relative abundance and the effects of fire on each species at each site. Due to the current lack of standardized soil sampling protocols in the ecological literature, and uncertainty around the definition of what denotes a fertile or infertile soil, I propose a number of standardized protocols and sampled according to these established protocols in order to accurately determine the soil nutrient status at each site. Following this, the relationships between climatic variables and soil nutrients with both species means and community weighted means for eight key leaf traits were explored. Although some significant relationships were found between savanna leaf traits of woody plants, climate, soil nutrients and their interactions, these tended to be weaker than those found in meta-analyses. These broad-scale studies usually include sites from many biome types, many of which are from temperate regions where inherent levels of disturbance are typically much lower than in African savannas. The high levels of disturbance typically found in African savannas are thought to partially account for the high within site variability found in leaf traits and the weak relationships found between leaf traits, soil nutrients and rainfall. To assess the importance of resources vs. disturbance in savannas functioning, the effects of soil nutrients, rainfall, fire and both meso-browser and mega-browser impact on twenty savanna woody plant traits relating to plant palatability, chemical and structural defences were explored. Structural defences were found to be more strongly correlated with soil characteristics than chemical defences, while browser impact was found to be strongly correlated with structural defences but not with chemical defences. Actual browser utilisation tended to be more predictable for meso-browsers than mega-browsers. Secondly using an experimental approach, two sets of herbivore exclosures were utilized to directly test how mammal browsers influenced woody species distributions, abundance, population structure and plant traits relating to palatability and defence. The effects of three longterm herbivore exclosures in the Kruger National Park on savanna woody plant community compositions, population demographics and densities were determined. Browsers were found to have significant impacts on species distributions, densities and population structures by actively selecting for species with favourable traits, particularly higher leaf N. An interaction between browsers and fire which limited the recruitment of seedlings and saplings into larger size classes was also demonstrated… [etc]

No description available.

Orientador: Anna Diva Plasencia Lotufo / Resumo: This work analysis load data from desegregated levels that presented difficulties to load forecasting with several methods due to variation in electrical energy consumption. The application proposed in this work is short-term load forecasting to a university building by GRNN (General Regression Neural Network) considering the bottom up approach and using a moving average filter to deal with the missing or wrong data. It is presented the system that provides the data as well as the methods used for pre-processing and realize the forecasting. The results are evaluated by MAPE (Mean Absolute Perceptual Error) and are considered good when compared with other methods. / Mestre

It is in the interest of organizations to understand that the public demand for quick and easy accessibility is growing at a rapid pace. However, the transition from traditional marketing strategies to social media marketing is not a straightforward process, but more complex.The aim of the thesis is to, from a bottom-up perspective, draw conclusions from a comparative study of the social media praxes of Umeå-situated businesses and organizations.The study indicates that the transition into social media still continues, but that there has been a change in organizations’ mindsets. It also shows that the application of a bottom-up approach provides new ways of viewing and understanding activities within social media.It is in the interest of organizations to understand that the public demand for quick and easy accessibility is growing at a rapid pace. However, the transition from traditional marketing strategies to social media marketing is not a straightforward process, but more complex.The aim of the thesis is to, from a bottom-up perspective, draw conclusions from a comparative study of the social media praxes of Umeå-situated businesses and organizations.The study indicates that the transition into social media still continues, but that there has been a change in organizations’ mindsets. It also shows that the application of a bottom-up approach provides new ways of viewing and understanding activities within social media.

Magister Scientiae - MSc / Our main results are that splay trees are faster for sorted insertion, where AVL trees are faster for random insertion. For searching, skip lists are faster than single class top-down splay trees, but two-class and multi-class top-down splay trees can behave better than skip lists. / South Africa