Academic journal article Human Factors

A Longitudinal Study of the Effects of Ecological Interface Design on Skill Acquisition

Academic journal article Human Factors

A Longitudinal Study of the Effects of Ecological Interface Design on Skill Acquisition

Article excerpt

INTRODUCTION

The vast majority of experimental evaluations of interface design concepts reported in the literature have been conducted over a short period, usually ranging from a few hours to, at most, 1 h/day for a few weeks. For several reasons, however, experimental results obtained over the short term may not generalize to the longer term. First, people may be able eventually to compensate for the deficiencies of a poor interface. If so, then performance disadvantages observed in the short term will disappear with practice. Second, differences in effort might also diminish over time. Thus if one interface is found initially to lead to less mental workload, one cannot be certain that this effect will persist over the long term, as the user's processing becomes more automatic.

Third, the full effects of skill acquisition may not be observable over a short period. This is especially true for complex human-machine systems; it may take some time before users can become well adapted to the demands placed on them. In this situation, one might find that differences between interfaces only appear over an extended period. Fourth, an interface that initially leads to "better" performance may actually lead to performance deterioration in the long run. For example, people could become overly dependent on the perceptual features of the interface and thereby exhibit a form of "deskilling" over time (see Wickens, 1992).

Therefore, there are many reasons to suggest that short-term evaluations of interface design concepts may generate potentially misleading experimental results. This is important from an applications perspective, given that interfaces for complex work environments will be used on a daily basis for years once they are introduced into the workplace. Thus there is a significant need to investigate the effects of interface design concepts over longer periods.

This paper describes a longitudinal study of ecological interface design (EID), a theoretical framework for designing computer interfaces for complex work environments. In the study, participants controlled DURESS (DUal REservoir System Simulation) II, a real-time, interactive process control simulation, for about 1 h approximately every weekday for a total of six months. To address the impact of interface design on skill acquisition, participants controlled the system with one of two interfaces: an interface based on the principles of EID containing physical and functional (P + F) system representations and a more traditional interface based solely on a physical (P) representation.

BACKGROUND

Theory

EID is a theoretical framework for designing interfaces for complex work environments. It is based on the skills, rules, knowledge taxonomy of levels of cognitive control (Rasmussen, 1983). The framework includes three prescriptive design principles, each directed at providing the appropriate interface support for a specific level of cognitive control. First, to support skill-based behavior, operators should be able to act directly on the interface. Furthermore, the structure of the displayed information should be isomorphic to the part-whole structure of movements. Second, to support rule-based behavior, the interface should maintain a consistent one-to-one mapping between the work domain constraints and the perceptual cues provided in the interface. Third, to support knowledge-based behavior, the interface should represent the work domain in the form of an abstraction hierarchy (Rasmussen, 1985), which can serve as an externalized mental model to support problem solving. This system model contains both physical and functional representations of the system.

Note that, in contrast to other interface design frameworks (e.g., Wickens & Carswell, 1995), EID provides guidance for identifying what information should be included in the interface, not merely for determining the visual form of the interface. See Vicente and Rasmussen (1990, 1992) for a more detailed description and justification of these design principles. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.