No description available.

Digital behaviour interventions are information systems that deliver advice and support to help people change their attitudes and/or behaviours. This thesis investigated the design principles that can make digital behaviour interventions more effective. In particular, this thesis focused on the design of dialogue systems aimed to encourage healthy behaviours. First, we explored the collected empirical evidence of past successful digital interventions to identify the best practices in designing such interventions. Based on these principles, we proposed a novel framework for creating personalised digital behaviour interventions. The framework emphasizes the process of personalisation at three levels: (1) modelling a personalised, dynamic model of behaviour of each user, (2) employing a personalised set of behaviour change strategies, and (3) personalising the delivery of each strategy to each user. Next, we looked at the effectiveness of a number of tactics that can be used by a system to enhance the delivery of behaviour change strategies. First, we investigated the conflicting results reported on the effect of onscreen characters on the system’s credibility. An onscreen character can enhance the system’s credibility if its appearance portrays high credibility with respect to the topic discussed by the system. For each topic, people do have a preference for which they would like to learn from. Finally, we combined all our findings to implement MARY, a virtual health trainer that encourages regular walking. MARY delivers a personalised intervention by supporting personalised goal setting, providing personalised daily feedback and emotional support. A longitudinal acceptability and feasibility evaluation of MARY with twenty-five participants confirmed our separate findings in a realistic setting. 90% of the participants improved or maintained their attitude towards walking, 81% of the participants intended to walk more, and 67% of the participants increased their daily step count.

A variety of traffic management schemes have been tried in the past to alleviate traffic problems, but with limited success. The European Space Agency proposed a satellite based Road Traffic Monitoring (RTM) system to overcome the limitations in existing schemes. This is studied in Aberdeen in the form of a simulation operating under varying conditions. This simulation has been introduced into the Internet using advanced Web technology to acquaint a wide range of users with the RTM concept. The work that has contributed towards this thesis looks at different approaches used for incorporating pre-processing and post-processing modules to strengthen the RTM simulation on the Web. A Database Management System was integrated into the system to improve data storage and manipulation tasks. This acts as the central coordinator to several sub-applications within the RTM system. A Graphical User Interface has been added to the front end of the simulator to provide easy access into the simulation through a sophisticated visual display. The post-processing needs are being handled by a spreadsheet suite. Techniques that were followed to integrate these modules with each other or with the core simulator have been considered in detail. The spreadsheet facilities have provided analytical ways and means of interpreting RTM system characteristics and performance criteria satisfactorily. As the thesis proceeds from the beginning to the end, it reflects how the original RTM simulator has evolved into a powerful simulation tool for analysing the effects and underling protocols of the RTM system comprehensively.

No description available.

No description available.

This research sets out to examine some of the implications of metaphor of the "vocabulary network" that is widely used in lexical research. It takes a formal approach to the exploration of this metaphor by applying the principles of Graph Theory to the notion of the lexical network. It looks, in particular, at how graph theoretical principles may be applied to word association data in order to compare the relative densities of L1 and L2 lexical networks. Nine experiments were carried out. An initial series of studies followed up in detail a simple graph theoretical model, proposed by Welsh (1988) and Meara (1992). Data was collected from learners and native speakers of French, English, Spanish and German using a word association chain-building methodology. These experiments showed up significant theoretical and methodological flaws in the original model which could not be readily circumvented by the development of a more sophisticated weighted network model. In view of this, an alternative elicitation technique was devised with the aid of a computerized simulation program. This alternative methodology was employed in a final experiment which showed it to be effective as a tool for the comparison of the densities of vocabulary networks in L1 and L2. The new elicitation technique was also able to provide a means of actually quantifying lexical density levels. It is argued that it will be helpful to develop a more complex interpretation of the notion of lexical density which includes parameters such as the inclusiveness of networks as well as the number of connections they contain. Our formal investigation of the network metaphor also suggests that we need to be more precise about the type of network model that is most appropriate to vocabulary research. It is proposed that L2 word association behaviour may be best predicted by semi-random, "small-world" network models, rather than by the wholly random networks that much of the literature tacitly assumes.

Multimedia - the integration under computer control of text, still and moving images and sound, as well as the integration at another level of broadcasting, publishing and information technology - has already begun to change, and is doubtless about to change even further, the way in which information is presented, whether for education and training, business or pleasure. As with every new form of communication in its early stages of development, multimedia will inevitably go through a period of unexpected discoveries, but also of experimentation and consequent mistakes. One of the main aims of this thesis is to attempt to show, indirectly and directly, that the number of potential mistakes in the area of the use of sound in multimedia applications can be reduced by approaching the subject in an informed way, that is, by calling on the growing literature on the subject. Essentially, the thesis is therefore a critical review of the relevant literature, combined with some theoretical conclusions, and a practical guide to sound for the multimedia designer/producer. This practical guide is developed fully in the final chapter, after looking in detail at speech, music, sound effects and some other factors bearing on the subject.

The first phase of experimentation studied user's perception of icons. When first encountering concrete, or pictorial, icons, users can compare the icon to their knowledge of the real world to deduce its meaning. This is much less likely for abstract icons, which usually consist of shapes, arrows and lines. A close relationship between icon concreteness and icon meaningfulness would therefore be expected. As users gain experience with icons, however, icon meaningfulness would therefore be expected to increase whilst concreteness remained constant. Experimentation found this to be the case. Consequently, although the icon characteristics of concreteness and meaningfulness may initially be closely related, with experience users perceive them as being independent icon characteristics. The second phase of experimentation examined the influence of icon characteristics , user experience, and mental workload on performance. The icon characteristics of concreteness, meaningfulness, familiarity and semantic distance were found to have considerable influence on user performance. Despite the emphasis on concreteness in previous research, semantic distance was most influential in determining novice users' performance. This was possibly because they relied on 'goodness-of-fit' between icons and their functions, rather than simply pictorialness, in order to learn icon-function relationships. Experienced users' response times were largely determined by icon familiarity, which appeared to act as an index of how easily representations could be accessed from long-term memory. Familiarity was also particularly important for high workload conditions. This research shows the importance of cognitive icon characteristics in determining user performance. Finally, the characteristics of primary importance were found to vary in accordance in experience and workload.

The visual interface presents the user with the opportunity to process and it makes sense of information. One of the most common and efficient forms of interacting with such a system is through direct dialogue with the icons presented on the interface. Although the advantages of these icons have received little attention in the past, recent studies are now beginning to assess the importance placed on the visual and cognitive properties they possess. The experiments outlined in this research examined the effect which a variety of visual cues could enhance users understanding and learning of icons. Of particular interest was the role which category cluster of icons could be identified visually for the user to help them learn the meaning of icon sets. Each experiment examined the role of a specific cue (or combination off cues). These studies revealed the importance of the visual attributes on the interface in providing implicit category cues to allow the user to learn the interface more easily and enhance usability. These findings are discussed in the terms of the users’ mental models of the interface.

No description available.