Evaluating immersive learning

Two decades ago, I was playing Tetris on my computer. The game issued one block at a time for me to orient before it fell into place, filling in the growing pile without gaps. Because my movements were so limited, the game could learn my habitual reactions and then outsmart me. It would issue an unexpected block that was also turned in ways I was not ready for. A decade later, Macromedia offered add-ons to Flash and Dreamweaver software for so-called learning. The user's movements were limited to dragging objects within a frame and inserting text into fields. The accuracy and time taken could both be scored.

This week, I reviewed some SCORM compliant software that supports avatars moving around in virtual space and conversing with programmed characters. Once again, the accuracy of movements, object selections and text entries could be scored, as well as counting mistakes and the amount of time taken. In spite of all the maneuverability of each learner's avatar within the virtual environment, the learning was as confined as if Java scripted objects were sliding around on a web page.

In pondering this experience since, I've realized how spellbinding the measurement of learning must be. We assume that measurable learning is what counts at the expense of intrinsic, serendipitous and social modes of learning. This spell induces uses of immersive environments for learning to pass up the rich opportunities for locating learning in space.

I suspect this pattern will persist until we rely on synchronous encounters with peers and coaches who's avatars observe our avatars in action. "Learning to evaluate others" will become part of the instructional objectives. We will get practiced at noticing the conduct, choices, confusion, and repetitive movements we can see in other avatars. We can also interact with other avatars in conversation and collaborative efforts. This will give us impressions of comprehension, competencies and ongoing curiosity they bring to the issues or skills under development. Evaluation of others' learning can then involve 360 degree feedback models, instead of quantitative measures of accuracy and duration of efforts. Individual bias will get diluted and compared by the variety of viewpoints included in an assessment.