"Cognitive, Social, and Cultural Needs of Learners Tell Me a Lot About Designing Software to Help Them Learn, or My Needs Analysis Begins with the Literature"
Janet Kolodner
Learning by Design Group
Interactive and Intelligent Computing
College of Computing

12:00 Noon on Thursday, October 26, 2006
ATDC, Centergy Building, Hodges Room 335

By custom, in HCI, needs analysis has centered on observations and ethnographic study of the situation(s) software is being designed for. What tasks are people carrying out? How do they do those things within their environment? In what ways can the computer help? In my work of the past 15 years, I've also been creating interactive computer systems, first for supporting design work, and more recently for supporting learning. In each case, I've conducted my needs analyses quite differently than is usual in HCI. My needs analyses begin with the literature. What do we know about how people design? What do we know about what is difficult for novice designers? What do our cognitive theories propose that might allow them to design more like experts? What do we know about how people learn from experience? What do we know about what's difficult about learning from experience? What does the literature tell us about mitigating some of those difficulties? What does all this suggest about designing the learning environment, the functionality our software might have, and the ways t hat software might be integrated into the life of the people in the environment? The literature, especially the cognitive literature, gives me my first clues about what is needed. I'll walk through the design of several computer systems and learning environments I've designed with my students, post-docs, and other colleagues over the past decade, making clear for each what the literature t old me about needs. Then I'll show, for each, what we designed as a result and discuss how observations and ethnographic studies took us to the next step. Such an approach, I think, will be more relevant and appropriate when engaging in human-centered computing design than the more traditional bottom-up approach.