NSERC Discovery Grants

Personalization of interaction based on discovered constructs (2012-2016)
Eyewitness identification is often highly regarded by the legal system, yet it is not always accurate. An identification may be compromised if the words used to describe the target face impair the witness' memory of that face (verbal overshadowing) or if the witness sees so many photos that he or she may identify a photo as the target face when it is actually similar to one of the early photos viewed (inaccurate source attribution). Faces of a race different than the witness may cause additional difficulties (cross-race effect). In order to improve the accuracy of eyewitness identification, these sources of inaccuracies must be mitigated. Verbal overshadowing may be addressed by using only non-verbal means to ask for recognition, rather than recall, of the target face. Inaccurate source attribution may be ameliorated by reducing the number of candidate photos presented to a witness. In order to meaningfully achieve this reduction, it is necessary to understand how a particular witness judges similarity between photos so that a computer-based presentation can be personalized accordingly. My research to date indicates that people employ one of at least two different strategies when judging facial similarities. I seek to democratize accurate identification of faces by: understanding the different strategies that people may use to judge facial similarity; developing a concise means of testing which strategy a witness is likely to apply; developing a computer interface that makes use of a witness' identified strategy to allow accurate identification of the target face, while viewing as few candidate photos as possible; and developing an extensible system to manage and store candidate photos for use with the interface. My research to date has used a set of 356 photos (178 Caucasian and 178 First Nations to allow for examination of cross-race effect), which participants have sorted based on how they each judge similarity. This work will contribute to eyewitness identification by providing better tools for witnesses and by providing a basis for understanding which faces may be more or less easy to identify. The same techniques can be used to study interaction with fractal images, which has applications for democratization of multimedia composition.


Democratization and personalization for information access (2006-2010)
Computer software should conform to its users and not the other way around. Although this phrase is often repeated, it seems to carry little weight today: users are regularly told of the need to use standard processes and software. My research explores an alternative view: let software deal with each user on their own terms and then let it manage the translation of each user's work into the terms of the others with whom he or she needs to collaborate. Democratization means that any user will be able to have satisfying and meaningful interactions and explorations of their chosen problem domains. Personalization means that each user is able to interact with the tool in a way that mirrors his or her own concept of the problem. Knowledge acquisition techniques will be used to discover each user's concept of a problem without requiring its explicit naming with a possibly unfamiliar vocabulary. User experiences will be more satisfying, hence more helpful to democratization, if the personalization phase can be as short as possible. I have developed applications in the areas of information visualization, recipe browsing, environmental decision support systems, eyewitness identification, and multimedia composition. I am seeking techniques to understand, as directly and as efficiently as possible, how any given individual conceives of a problem. I plan to take this understanding or knowledge and make it available for manipulation by that individual through my software, called cogito. This interface is intended to allow and support a user's exploration of that information, enabling him or her to manage and even transform the space (by changing the order in which decisions are made, for example) until the desired results are obtained. People will bring their diverse applications or problems to the cogito system and they will have satisfying experiences exploring them. Technology can, I believe, be more powerfully used when it augments individuals seeking personally-creative solutions and helps those individuals to communicate those ideas to others.


Conceptual spaces: theory and tools (2002-2005)
In 1960, J. C. R. Licklider looked to a time when humans and computers would work in intimate association and saw it as being "intellectually the most creative and exciting in the history of mankind." Back then, Licklider found that up to 85% of his "’thinking’ time was devoted mainly to activities that were essentially clerical or mechanical ... operations that can be performed more effectively by machines than by men." Although many of these tasks are now done by computers, many software programs still keep their users from their work by requiring them to be tooinvolved in clerical or mechanical details. In an era when so much is done by computers, humans are still needed to deal with the problems that require their personal creative input. These problems are harder to deal with, and as such only experts are likely to tackle them. My research focuses on developing appropriate computer-based tools to support the experts in working on these problems, and to encourage more participation from more people in exploring and evaluating alternative solutions to these hard problems. The tools in which I am particularly interested use visual interfaces to remove the requirement of computer expertise from their users and to explore and evaluate all the available solutions. Such interfaces address the problem when a user, searching for a solution, says "I’ll know it when I see it" but has no other way to specify it. The notion of a conceptual space, defined with the help of a programmer, is a way for the user to build a vocabulary to talk about his or her problem. In this way, my work seeks to change the relationship between users and programmers rather than put programmers out of work. To develop these tools, I will rely on the study of software users to refine my own software and ideas. Widespread acceptance and use of this type of tool will go a long way to realizing Licklider’s vision, meaning that computer use will become more productive and more rewarding.