Tuesday, July 26, 2005

The Learning Object Paradox

Parkin's Learning Object Paradox (PLOP) states that: "The more reusable a learning object becomes, the less useable it is." This is because the usability of a learning object varies in direct proportion to its size while its reusability varies in indirect proportion to its size.

Think in terms of bricks, rooms, and houses. Bricks can be interchanged without affecting the harmony of a house design. Rooms cannot. The smaller your learning objects become, the easier it is to slip them in to other uses without creating any major disruption, but the less “meaningful” they are. The larger the objects become, the less re-usable they get, because they become more context-rich. But you get to a point where the size of the object is large enough to be self-fulfilling and truly meaningful, usually at the level of a house, or whole course.

Which is why, in practice, most learning objects are no smaller than a course. A course is not very re-usable, though you may fit it into different curricula, in the same way that universities fit different courses into different degree programs.

You don’t hear much about learning objects and sharable content objects any more, at least not in mainstream training circles. But there is still an enormous effort going on among learning technologists to make this idea more workable. Are these efforts rather like trying to build a better steam engine long after the internal combustion engine has gained popularity?

The original concept of sharable content objects (in a training context) assumed that learning is primarily content-driven, and that learning content can be decomposed into smaller and smaller components that still retain some inherent independent pedagogical value. If your objects are designed to “click” together, like a child’s construction toy, you can create lots of different learning experiences by clicking together the appropriate objects from your repository of already-created content. The argument went that while the initial cost of building a decomposable course might be higher than building a stand-alone course, the ultimate savings derived from reusing content would be significant.

The historical limitations of learning objects (particularly as manifested in early SCORM) have resulted in dire canned e-learning course design, or have led instructional designers to simply wrap a SCORM interface around whole courses. Today, the “object” is rarely smaller than an entire course. Effectively, interoperability of courses – the ability to run a course on any conformant LMS – has been the primary benefit of SCORM, rather than the reusability of smaller content objects.

The notion of reusable learning objects is in many ways counter to the web ideal of dynamic personalized content, the ISD ideal of internally consistent look, feel and sound to learning flows, and the pedagogical ideal of performance-objective driven custom-built learning experiences.

Re-usability implies that a learning object developed for use in one context can be simply plugged into another context. It also suggests that such an object ages slowly, staying relevant long enough to be reused often enough to justify its reusability. This is more true in some fields than in others. So, for example, you develop an object (say a lesson describing Maslow’s Theory) as part of a management course on motivation. Later, you develop a course on applied psychology and you don’t need to re-create the Maslow lesson because you can simply lift it from the earlier course and drop it into the later course. As the theory itself never changes, the object can be expected to have a long life -- so long as its content is devoid of context, which is typically pretty mercurial.

We have all taken short cuts in the classroom, pulling slides from one session and using them in another, even if they have different layouts or color schemes. That may work better in a classroom setting than online, because there is continuity in the primary medium: the trainer. But online, without the mediation of a live instructor, not only can you end up with all sorts of disconnects in terms of media (different voice, graphic style, look, fonts, pace, interactivity), you are also likely to get disconnects in terms of pedagogical approach. This is avoidable, but only by imposing content standards that are dull and uniform, so every course ends up looking, sounding, and running like every other.

Perhaps the most important obstacle to the success of sharable content objects is the fact that learning is not primarily about content, or about courses. Those who glibly pronounce that “content is king” really irritate me, because it is patently untrue. While content is obviously essential, context and process are more important to learning. But that’s a rant for another day.

Despite all of the reservations and difficulties, the idea of reusability has enduring appeal. But today, as collaborative community-based learning starts to take shape, perhaps we should be re-thinking what a learning object might be. Instead of looking at a learning object as a chunk of easily-connectable content residing on a server, maybe we should be looking at an easily-connectable person residing on a network...


Original in TrainingZONE Parkin Space column of 8 July 2005

1 comment:

Adam said...

I couldn't agree more. For years I struggled to understand SCOs and try to make them work...to dismal failure. Context truly trumps content in meaningful learning designs and this can't simply be decomposed into its parts (gestalt theory here?).

On the other hand, the advent of wikis, blogs, RSS, and other "web 2.0" technologies may be able to breathe new life into this idea. An index linking to various pages in a wiki and blog posts (each of them could be considered a "learning object", no?) may create an effective course.