Course 33: Cognitive Crash Dummies: Predicting Performance from Early Prototypes -
CourseContribution & Benefit: Presents a free tool that integrates rapid UI prototyping with predictive human performance modeling. Participants use their own laptop, learn to mock-up interactive systems, and create models of skilled performance.
Abstract » Prototyping tools are making it easier to explore a design space so many different ideas can be generated and discussed, but evaluating those ideas to understand whether they are better, as opposed to just different, is still an intensely human task. User testing, concept validation, focus groups, design walkthroughs, all are expensive in both people's time and real dollars.
Just as crash dummies in the automotive industry save lives by testing the physical safety of automobiles before they are brought to market, cognitive crash dummies save time, money, and potentially even lives, by allowing designers to automatically test their design ideas before implementing them. Cognitive crash dummies are models of human performance that make quantitative predictions of human behavior on proposed systems without the expense of empirical studies on running prototypes.
When cognitive crash dummies are built into prototyping tools, design ideas can be rapidly expressed and easily evaluated.
This course reviews the state of the art of predictive modeling and presents a tool that integrates rapid prototyping with modeling. Participants will bring their own laptops and learn to mock-up an interactive system and create a model of skilled performance on that mock-up. The course ends with a review of other tools and a look to the future of predictive modeling.
AUDIENCE
Designers, usability professionals and software developers who want to evaluate alternative designs alternatives. No prior knowledge of prototyping, psychology or predictive human performance modeling is required.
INSTRUCTOR
Dr. Bonnie E. John has more than 25 years experience in HCI. A CHI Academy member now at IBM Research, Dr. John was head of CMU's Masters Program in HCI for a dozen years, researches both human performance modeling and software engineering, and has consulted regularly in government and industry. She has taught CHI courses since 1992.