Although concepts are located within individual minds, while word forms are shared across entire language communities, words and concepts are normally deemed to be tightly bound. But in fact, at least to the extent that concepts vary, the relationship between words and concepts may not be as uniform or stable as is often assumed. Nevertheless, language may itself mediate that relationship, through its entrenchment and use. Psychologists have already investigated language use in referential communication, but they have yet to focus in detail on the role of language in conceptual coordination. One of the obstacles has been the theoretical and methodological challenges that arise from seriously abandoning conceptual universals. To that end, an experimental framework was developed based on sorting tasks in which participants freely partition a set of stimuli into categories and an objective measure for comparing two outputs. Four experiments were then conducted to investigate whether people were conceptually coordinated before, during and after linguistic interaction. Experiment 1 consisted of a cross-linguistic study looking at default coordination between native speakers. Participants both sorted items into groups and named them individually. There was a relatively high degree of categorisation agreement among speakers of the same language, but not nearly as high as for naming agreement. Experiments 2-4 inquired into conceptual coordination during or immediately after linguistic interaction. Experimental manipulations involved the form of language use (full dialogue or only category labels), as well as the type of feedback (category groupings, labels, both, or neither). In particular, Experiment 2 investigated the effects of categorising a set of objects together, with or without dialogue, on subsequent individual categorisation. The results were inconclusive and revealed specific methodological issues, but yielded interesting data and were encouraging for the general framework. Experiment 3 modified the designwhile testing and extending the same general hypotheses. Participants carried out a sequence of categorisation tasks in which they tried to coordinate their categories, followed by individual categorisation and similarity tasks. The availability of dialogue and feedback was manipulated in the interactive tasks. During interaction, they also received both kinds of feedback, except in the control condition. Pairs that could talk coordinated much better than the others, but feedback didn’t help. Experiment 4 looked into the effects of the four possibilities for feedback during a longer sequence of interactive tasks. In general, conceptual coordination was found to depend on grouping feedback only. However, by the end of the task, pairs who received both kinds of feedback did best. All three interactive experiments also measured lexical convergence between pairs. The results generally revealed a dissociation, with lexical alignment showingmore convergence and occurring under a wider variety of conditions. Togetherwith previous research, these findings showthat language can bring about conceptual coordination. However, it appears that the richer the form of language use, the more conceptual convergence occurs, and the closer it gets coupled with lexical convergence. The long-term effects, if any, are much weaker. These studies have implications for the general role of language in cognition and other important issues.

The present research attempts to develop a general model of the diffusion of Western loanwords in contemporary Japanese within the variationist framework. It describes and predicts, based on empirical evidence from apparent- and real-time data, the elaborate process of changes in favor of loanwords as opposed to their existing native equivalents. First, people's self-reporting shows a consistent tendency for a younger generation to show a stronger preference for loanwords than an elder one. This indicates changes in favor of loanwords are in progress in apparent time (Chapter 4). Second, the above-mentioned age gradient is attested to by corpus-based data. It also reveals that the occurrence of loanwords is accounted for multi-dimensionally by a wider range of language-external factors such as generation, education, register and style (Chapter 5). Third, an in-depth study of the individual loanword keesu (< case) reveals that not only external factors but also internal ones, e.g., usage and collocation of the word, have impacts on its occurrence (Chapter 6). Fourth, an investigation of the loanword sapooto (< support) shows that a stylistic variable comes into play in its diffusion in interaction with an educational variable. The loanword is disfavored when the speech setting shifts to formal in particular by the most educated speakers (Chapter 7). Fifth, a real-time approach to loanword adoption verified that individuals can change their language attitude or behavior throughout their lifetime. It highlights importance of longitudinal observation of the phenomenon in making a more accurate prediction of change (Chapter 8). The present research confirms that the occurrence of loan variants is bound by various social and linguistic contexts. The above empirical findings contribute to the field of variationist study by opening up the possibility of analyzing linguistic variation in Japanese at the lexical level.

Fechamento de loops é um dos principais processos das estratégias de SLAM baseadas em grafos, usadas para estimar o erro de deslocamento acumulado à ser minimizado pela técnica. Neste sentido, boas correspondências de cenas permitem criar uma conexão entre dois nós do grafo que está sendo construído para representar o ambiente. Contudo, falsas correspondências podem levar essas estratégias a um estado irreversível de falsa representação do ambiente. Neste trabalho, um método robusto baseado em features que usa sequências de imagens para reconhecer áreas revisitadas é apresentado. Este método usa a abordagem de Bag-of-Words para reduzir efeitos de iluminação e uma ponderação TF-IDF para ressaltar as principais features que descrevem cada cena. Além disso, um algoritmo baseado na técnica de Mean Shift é usado sobre uma matriz de similaridade para identificar a possível trajetória seguida pelo robô e melhorar a detecção de fechamento de loop. O método apresentado foi testado em um ambiente aberto usando sequências de imagens coletadas com usando uma câmera de mão e um drone modelo Parrot ArDrone 2.0. / Loop closure recognition is one of the main processes of graph-based SLAM strategies, used to estimate the accumulated motion error to be minimized by the technique. Good scene correspondences allow to create constraints between two nodes in the graph that is currently being built to represent the environment that the robot is immersed. However, false correspondences can lead these strategies to an irreversible wrong environment representation. In this work, we present a robust feature-based loop closure approach that uses image sequence matching to recognize revisited areas. This approach uses Bag-of- Words to reduce the effects of lightning changes and a TF-IDF weighting to enhance the main features that describe each scene. Besides, an algorithm based on Mean Shift is used over a similarity matrix to identify the possible trajectory followed by the robot and improve the loop closure detection. Our method is tested in a GPS-denied outdoor environment using image sequences collected using a handheld camera and a Parrot ArDrone 2.0.

An increasing number of ebusiness projects are undertaken as voluntary collaborations of independent organizations. These projects are known as ‘ebusiness collaborative projects’. Industry sources estimate only about 20% of these projects achieve their stated objectives due partly to inadequate project management. The volunteer and virtual nature of the collaboration and relationships among the participating organizations means existing project management frameworks and methods are not appropriate because they assume project management has authority, or that projects occur within a single organization. This thesis investigates the nature of ebusiness collaborative projects to provide insight into management approaches that increase the likelihood of success. The research questions are: What concepts need to be addressed in a framework capable of supporting effective management and providing an understanding of ebusiness collaborative project outcomes? Is there an existing framework with this capability? Are the concepts supported by empirical evidence and a sound theoretical base? Can these concepts be used to form an integrated framework suitable for use by practitioners? A qualitative meta analysis of 6 published case studies identified characteristics that explained challenges specific to ebusiness collaborative project management. Empirical research identified concepts that need to be addressed in a management framework, namely: 3 organizational levels – organizations, teams and representatives; 3 stages of the eBusiness Collaborative Project Lifecycle© and; 4 management meta factors - motivation, capability, communication and coordination These concepts were supported by theory. In the absence of an existing framework, the concepts were used to develop the Integrated Framework for Managing eBusiness Collaborative Projects. Five additional case studies were used to evaluate the Integrated Framework and its usefulness for practitioners. Findings indicate the Integrated Framework is capable of supporting effective management and providing an understanding of the outcomes of ebusiness collaborative projects within the Australian context. This thesis contributes to knowledge by integrating and building on theory and existing research about project management, collaborations and virtuality and applying these findings to a real world environment. The Integrated Framework enhances industry best practice and may apply to all collaborative projects in which participating organizations volunteer to work towards an agreed objective or outcome.

Embodied cognition theorists suggest that cognition is bodily based and that the brain developed due to interaction with the environment, and thus evolved to facilitate sensorimotor processing. As such, one goal of embodied cognition research is to determine how the interaction between the body and the environment affects the storage and processing of semantic information. Recent neuroimaging research has shown that the sensorimotor and premotor cortices are activated somatotopically when responding to action-related stimuli. In addition, behavioural research has provided evidence in support of the theory of embodied cognition, in that the sensorimotor properties of a stimulus have been shown to affect performance on language tasks. The goal of the current research was to provide a novel and comprehensive examination of the theory of embodied cognition through the combination of multiple experimental paradigms. Several functional magnetic resonance imaging and behavioural experiments on healthy participants were carried out, as well as a behavioural study of two individuals who have undergone either a left or right hemispherectomy. The results from the functional neuroimaging experiments demonstrated that there are common regions of activation between motor movements and semantic processing, whereby sensorimotor and premotor regions that are responsible for arm and leg motor movements are also recruited when responding to arm- and leg-related action semantic knowledge. Thus these results are consistent with the theory of embodied cognition, suggesting that the motor system is involved in the processing of action-related semantic information. The behavioural results were also consistent with previous research showing that pictures have privileged access to action-related semantic knowledge. Additionally, the behavioural results with hemispherectomy patients provided evidence regarding the necessity versus sufficiency of the left and right hemispheres when responding to arm- and leg-related semantic knowledge. Finally, given that words rated as higher in body-object interaction were responded to faster than words rated as lower in body-object interaction, these results show evidence that language processing is grounded in bodily interaction and sensorimotor processing. Together, the results further advance the theory of embodied cognition, and moreover, provide an in-depth analysis of how arm- and leg-related stimuli are processed dependent upon presentation format.

Embodied cognition theorists suggest that cognition is bodily based and that the brain developed due to interaction with the environment, and thus evolved to facilitate sensorimotor processing. As such, one goal of embodied cognition research is to determine how the interaction between the body and the environment affects the storage and processing of semantic information. Recent neuroimaging research has shown that the sensorimotor and premotor cortices are activated somatotopically when responding to action-related stimuli. In addition, behavioural research has provided evidence in support of the theory of embodied cognition, in that the sensorimotor properties of a stimulus have been shown to affect performance on language tasks. The goal of the current research was to provide a novel and comprehensive examination of the theory of embodied cognition through the combination of multiple experimental paradigms. Several functional magnetic resonance imaging and behavioural experiments on healthy participants were carried out, as well as a behavioural study of two individuals who have undergone either a left or right hemispherectomy. The results from the functional neuroimaging experiments demonstrated that there are common regions of activation between motor movements and semantic processing, whereby sensorimotor and premotor regions that are responsible for arm and leg motor movements are also recruited when responding to arm- and leg-related action semantic knowledge. Thus these results are consistent with the theory of embodied cognition, suggesting that the motor system is involved in the processing of action-related semantic information. The behavioural results were also consistent with previous research showing that pictures have privileged access to action-related semantic knowledge. Additionally, the behavioural results with hemispherectomy patients provided evidence regarding the necessity versus sufficiency of the left and right hemispheres when responding to arm- and leg-related semantic knowledge. Finally, given that words rated as higher in body-object interaction were responded to faster than words rated as lower in body-object interaction, these results show evidence that language processing is grounded in bodily interaction and sensorimotor processing. Together, the results further advance the theory of embodied cognition, and moreover, provide an in-depth analysis of how arm- and leg-related stimuli are processed dependent upon presentation format.

In this thesis, we show that several natural questions about automatic sequences can be expressed as logical predicates and then decided mechanically. We extend known results in this area to broader classes of sequences (e.g., paperfolding words), introduce new operations that extend the space of possible queries, and show how to process the results. We begin with the fundamental concepts and problems related to automatic sequences, and the corresponding numeration systems. Building on that foundation, we discuss the general logical framework that formalizes the questions we can mechanically answer. We start with a first-order logical theory, and then extend it with additional predicates and operations. Then we explain a slightly different technique that works on a monadic second- order theory, but show that it is ultimately subsumed by an extension of the first-order theory. Next, we give two applications: critical exponent and paperfolding words. In the critical exponent example, we mechanically construct an automaton that describes a set of rational numbers related to a given automatic sequence. Then we give a polynomial-time algorithm to compute the supremum of this rational set, allowing us to compute the critical exponent and many similar quantities. In the paperfolding example, we extend our mechanical procedure to the paperfolding words, an uncountably infinite collection of infinite words. In the following chapter, we address abelian and additive problems on automatic sequences. We give an example of a natural predicate which is provably inexpressible in our first-order theory, and discuss alternate methods for solving abelian and additive problems on automatic sequences. We close with a chapter of open problems, drawn from the earlier chapters.

The first of these chapters was presented in 1904 as the author's Thesis (Ph. D.)--University of Pa. The material for the second was prepared by the author while Harrison Fellow for Research in the same University in 1904-1905. cf. p. 5.

Not available / text

published_or_final_version / Chinese / Master / Master of Philosophy