Dr. Kitajima has been a researcher with the National Institute of Advanced Industrial Science and Technology (AIST) since 1980. He was a research associate at the University of Colorado in Boulder in 1991. He received his Doctorate in Engineering from the University of Waseda in 1986, and a BS and MS in Physics from the Tokyo Institute of Technology in 1978 and 1980, respectively.

Kitajima started his research career at AIST (it was then called Industrial Products Research Institute) with the goal of developing a reading machine for the blind (1980 - 1985). He developed a novel method for recognizing printed Japanese characters. He then implemented the method with a prototype micro-computer-based reading machine that was capable of converting printed characters, captured by scanning a hand-held CCD camera, into synthesized voices at a speed of up to four phonemes per second. He earned his doctorate of this project in 1986 from the University of Waseda as a result.

His research interests then shifted to cognitive processes in human-computer interaction. He developed a formal representation system for human-computer interaction processes, that is used to make quantitative predictions about human performance, such as working memory load, ease of knowledge transfer, and so on.

Kitajima's current research is on how information contained in displays supports skilled performance and guides learning by exploration and deliberate search. In collaboration with Professor Polson of the Institute of Cognitive Science, University of Colorado in Boulder, USA, he has developed a computational model of skilled use of an application with a graphical user interface that explains errors made by experienced users. The model is based on Hutchins, Holland, and Norman's (1986) analysis of direct manipulation and is implemented using Kintsch and Mannes's (1991) construction-integration theory of action planning.

The action planning model has been expanded to include processes for comprehending instructions. The expanded model is called the LICAI Model. LICAI is an acronym for the LInked model of Comprehension-based Action planning and Instruction taking. When LICAI is pronounced [li kai], the pronunciation is the same as the Japanese two-kanji word,

which means comprehension. Kitajima is also applying this model to the evaluation of experimental displays for the autoflight systems of modern commercial aircrafts.

A new model currently being studied is the model that simulates users' navigation through the Web. This model is called CoLiDeS, a Comprehension-based Linked model of Deliberate Search. An important extension to the LICAI model is the inclusion of attention mechanism that enables the model to focus on a part of information abundant display for successful navigation (demo of the CoLiDeS model).