Guidance in the Interface: Effects of Externalizing Information During Problem Solving

Abstract

How can we design technology that suits human cognitive needs? In this chapter, we review research on the effects of externalizing information on the interface versus requiring people to internalize it. We discuss the advantages and disadvantages of externalizing information. Further, we discuss some of our own research investigating how externalizing or not externalizing information in program interfaces influences problem-solving performance. In general, externalization provides information relevant to immediate task execution visibly or audibly in the interface. Thus, remembering certain task-related knowledge becomes unnecessary, which relieves working memory. Examples are visual feedback aids such as “graying out” nonapplicable menu items. On the contrary, when certain needed task-related information is not externalized on the interface, it needs to be internalized, stored in working memory and long-term memory. In many task situations, having the user acquire more knowledge of the structure of the task or its underlying rules is desirable. We examined the hypothesis that while externalization will yield better performance during initial learning, internalization will yield a better performance later. We furthermore expected internalization to result in better knowledge, and expected it to provoke less trial-and-error behavior. We conducted an experiment where we compared an interface with certain information externalized versus not externalizing it, and measured performance and knowledge. In a second session 8 months later, we investigated what was left of the participants’ knowledge and skills, and presented them with a transfer task. The results showed that requiring internalization can yield advantages over having all information immediately at hand. This shows that using cognitive findings to enhance the effectiveness of software (especially software with specific purposes) can make a valuable contribution to the field of human-computer interaction.