Review by: Reichert, Raimond (2004-07-01)

When it comes to creating the metadata on any kind of (learning) object, the types of the metadata items must be succintly defined. Schulmeister addresses metadata standards such as European Ariadne standard or IEEE’s LOM standard, and focuses on the metadata item of human-computer interactivity of multimedia components as found in interactive learning environments. The standards merely provide a scale from “very low” to “very high” interactivity. He argues that such a rating depends too much on the subjective viewpoints of the metadata author, and in this paper outlines a qualitative framework of interactivity categories that can be applied to multimedia components such as diagrams, animations, or Java applets.

Schulmeister’s basic definition of interactivity is the amount of freedom of action users have in using the multimedia component. His model of interactivity defines six levels of increasing human-computer interaction. Level one means no interaction at all, but only a display of information, that is, there is no interaction on this level. Level two lets users navigate through the representation of information, choosing which links to follow. Level three offers multiple representations of the same content. Even on this level, there is hardly any interaction, but it provides the user with a choice of representation. On level four, the user can modify parameters of the representation. For example, in a 3d view, the user could move and rotate the object or adjust the lightning as needed. Additionally, on level five, the user can manipulate the content itself. The content is pre-fabricated, and the user can modify it to experiment with it. Interactivity level six means the user can create and manipulate content objects and watch the system react. This level provides the users with the highest degree of freedom in interacting with the multimedia component.

This model of interactivity is easy to understand and has the advantage of clearly distinguishing the increasing levels of interaction, from manipulating the representation to manipulating the content itself. What is less clear is the distinction between multimedia component interaction and program component interaction that Schulmeister makes in the second part of the paper. The distinction seems artificial, the boundaries blurred: Multimedia components can be regarded as special instances of program components. The article closes with a call to investigate the often-heard hypothesis that higher levels of interactivity have a positive effect on learning efficiency and learning results.