Investigating user knowledge transfer and its implication on product design

摘要

This paper investigates users' knowledge transfer performance on two email system platforms with different transparency levels and task change types. The objective is to provide guidance regarding task design in product upgrading and product extension. Four types of transfer tasks (commission, omission, sequence and no change) and two levels of transparency (high and low) are introduced in the study to evaluate users' performance in knowledge transfer. The paper hypothesizes that commission and omission tasks will realize the highest levels of performance time per step (PTPS) and error ratio. High transparency tasks will realize lower PTPS and error ratio than low transparency tasks. Two software packages, Camtasia and U-log, were used to record user input. The results show that commission tasks lead to the greatest increase in both performance time per step (PTPS) and error ratio, while performance was best during sequence tasks in which they had a positive transfer in PTPS and no increase in error ratio. Subjects also demonstrate better knowledge transfer in higher level of transparency tasks than in lower level of transparency tasks. In addition, the paper examines learning curves of the user performance. The majority of the users' performances in knowledge transfer (PTPS and error ratio) follow a typical learning curve. Abnormal learning curves are also analyzed. The impact of expertise on learning curves was studied. Expertise is quantified using a questionnaire about subject experiences. The paper concludes that in future design of product changes and the subsequent knowledge transfer, task change type and task transparency level is critical. A refined design approach is necessary to ensure that the design tasks are highly transparent to users. Additionally, the analysis of users' learning curve would provide reference for determining the threshold of users' adaptation to the new product, which is critical in consumer product design and user training.