The purpose of this chapter is to introduce an instructional technology concept known commonly as the “learning object.” First a review of the literature is presented as groundwork for a working definition of the term “learning object.” A brief discussion of instructional design theory is followed by an attempt to connect the learning objects approach to existing instructional design theory, and the general lack of such connective efforts is contrasted with the financial and technical activity generated by the learning objects notion. The LEGO metaphor frequently used to describe learning objects is critically examined and a successor metaphor is nominated. A taxonomy of learning object types is presented as a foundation for continued research in learning objects and related instructional design theories. Finally, the connecting of instructional design theory to the taxonomy is demonstrated and the benefits of this approach are briefly espoused.

This book section serves as an introductory chapter to the book on instructional use on learning objects. It defines the term learning object and proposes a metaphor for and a taxonomy of learning objects. The fundamental idea behind digital learning objects is described as enabling instructional designers to build relatively small instructional components that can be reused in different learning contexts.
There are many definitions of the term learning object. The definition by the IEEE’s LTSC, for example, is deemed too broad, failing to exclude anything. Another definition by Educational Objects Economy is deemed too narrow, as it includes only Java applets as learning objects. The author proposes the following definition: “any digital resource that can be reused to support learning", which is a proper subset of the LTSC’s definition.
Two major issues regarding learning objects are discussed from an instructional design point of view, combination and granularity. With regards to combination – or sequencing, to use an instruction designers’ term – the author notes that the standards discussion has largely ignored instructional design issues, even though promises of “automatically composed lessons" are frequently made. With regards to granularity, the big question is the appropriate size or scope of a learning object. This is an issue which does not have a clear answer.
The author proposes a successor metaphor, atoms, for learning objects to the often used metaphor of LEGO blocks. The main differences are: Atoms can not be combined arbitrarily; they can only be assembled in structures prescribed by their own internal structure; and it is not easy to assemble atoms, that is, training is required. This contrasts with the idea of having LEGO-like learning objects that can be treated like information objects in content management systems.
A taxonomy of learning objects which differentiates between five types is proposed, with the following examples to illustrate each type:

1. Fundamental objects, e. g. an image of someone playing a chord on a piano.
2. Combined-closed objects, e. g. a video with audio of someone playing a chord on a piano.
3. Combined-open objects, e. g. a web page combining the image and video with text material.
4. Generative-presentation objects, e. g. an interactive program which presents users with chord identification problems.
5. Generative-instructional objects, e. g. an environment where users are instructed on piano playing and can practice as well.

The main point of the paper is that instructional design theory must be incorporated in learning objects implementations that aspire to facilitate learning. The above taxonomy can help in providing prescriptive guidance of the type “for this type of learning goal, use this type of learning object".
However, the underlying definition of learning object does not exclude mere information objects. This is evident in type (1) and (2) of the taxonomy, and even type (3). Maybe it would be helpful not only to require that an object supports learning – many information objects do that – but rather that an object be connected to a learning goal. Then a clear distinction between information objects (type 1, 2, 3) and learning objects (type 4, 5) could perhaps be made.
Source: http://www.elearning-reviews.org/