T4: A Cognitive Approach to Interactive System Design

Full day tutorial for NordiCHI 2002 in Aarhus in October 2002.

Michael Atwood, Drexel University's College of Information Science and Technology


You will learn how to successfully design and improve useful and useable interactive systems by being able to answer the following questions.

  • What makes cognitive task analysis difficult? What can be done to alleviate this problem?
  • How do the users of the system think about their tasks and work they do?
  • How can I tell how useful and useable a system is now or how useful and useable it can be?
  • How do I get started in designing a cognitively useful and useable system?
  • How can I determine what parts of a system should be changed and how to change them?
  • How can I communicate well with others on my design and development team?


A successful start to designing useful and useable systems involves three phases:

  1. Deciding what you need to do. The first phase in design, deciding what you need to do, is easy if you ask the right questions. Surprisingly, these questions are often overlooked. Failing to ask these questions will very likely result in a design disaster. From a cognitive perspective, the right questions to begin with are who is going to use the system, what are they going to use it for, what do they know or think about the work they perform, what systems they will work with, and what aspects of the environment are crucial.
  2. Doing it. Once a designer has decided what they need to do, they need a way to represent the design. For this. We introduce the Goal-Oriented Model (GOM) representation which combines the cognitive focus of GOMS with the user-centered focus of the Scandinavian school of design.
  3. Evaluating how well you did it. The goal is "getting it close the first time," not "getting it right the first time." We introduce the Cognitive Walkthrough as a useful tool for this phase.


This tutorial is intended for anyone who is or who will be part of a team that designs and develops interactive systems to support complex human work or problem solving. The objectives of this tutorial are to provide attendees with the skills needed to ask and answer questions about how the people who will use the system think about their tasks and their work.


Brief lectures, discussion, as well as "minds on" and group exercises.


Dr. Atwood joined the faculty at Drexel University's College of Information Science and Technology as a Professor in the Fall of 1998. His teaching and research interests include human-computer interaction, computer-supported cooperative work, and artificial intelligence. Prior to joining Drexel, Dr. Atwood worked in industry, primarily as a manager of several research and development groups that successfully used the development of useful and useable systems as the basis for research programs. Dr. Atwood's earliest publications were influential in defining the field now known as human-computer interaction and he continues an active role in developing the field of HCI and the international HCI community.

Last update: 28. September, 2002 by webmaster@www.nordichi.org