DesiL (Designing for Learnability): a non-deterministic methodology for informing the design and assessment of easy-to-learn microcomputer interfaces



Journal Title

Journal ISSN

Volume Title


Texas Tech University


This study serves to commence research on a new concept within the purview of human-computer interface design. This concept is called learnability and addresses the ability of a computer user to learn how to use software without explicit training. The concept of learnability differs from the more common concept of usability in that it serves a purpose different from the "engineering type" of work performed by usability that focuses on contributions from cognitive science and human factors researchers. In particular learnability is concerned with factors local to the user's situation that may not be predictable a priori. While usability engineering aims to prescribe how human computer interfaces are designed, learnability may be thought of as a descriptive concept, that may only inform the design of computer interfaces.

Because this study seeks to begin research in learnability the research questions in this study are formative to assist in dimensioning the factors surrounding this concept and how they may be better understood and studied. The research questions permitted the discovery and exploration of factors affecting learnability and how those factors may then be designed into a human computer interface. The study also forwards several methodologies for assessing if a learnable human-computer interface has been achieved. This study is grounded in theories of social science, human communication and instructional design. However, it deviates from these theories in important ways when they were found to conflict with this study's definition of learnability. In particular, this study extends the work of Cool (1993), Carroll (1984, 1990, 1992) and Suchman (1987) in their framing of the problems encountered by persons both learning to use and using computers to accomplish everyday work. Their work, and this study conceives of learning-how-to-use-computers as a constructivist phenomenon that is effected by social, perceptual and interpretive factors ancillary to the actual learning task. This study's methodologies are therefore taken from social science research in symbolic interaction, ethnomethodology and the unconventional stochastic modeling of human communications processes.

When these methods were used to discover factors related to how a computer user perceives, interprets and performs work, and those discoveries used in informing the design of a human-computer interface, the research participants were found to be able to learn how to use the computer to use the underlying software without explicit training. It is therefore possible that these perceptual and interpretive factors play an important role in learning how to use computers. More research is needed to better understand the new concept of learnability and to dimension the factors that contribute to its successful implementation in the practice of human-computer interface design.