I really enjoyed last week’s class, a brief but valuable introduction into several learning theories.
A few observations: the theory of dual coding seems like it will necessarily run up against the limits of cognitive load, and I doubt there are any general principles that would guide a designer to know exactly where the line is between appropriately diverse multi-modal imprinting, and redundancy and overload. User testing is needed for each audience and content domain.
The theory of encoding specificity also grabbed my attention. It seems undoubtedly true; the more lifelike learning is, the more authentic the simulation, teaching environment, and curriculum, the more likely learning will transfer into “real life.” But of course, a teaching environment is almost always somewhat different from real life, otherwise you are just “learning-on-the-go,” which has its benefits, but doesn’t allow for low-risk mistakes and doesn’t allow for scaffolding. So the theory of encoding specificity really begs the question: in what way more lifelike? In other words, in what dimensions does a learning environment need to be lifelike, and in what dimensions is it better, or at least tolerable, for a learning environment to be simpler, lower-risk, or exaggerated, in contrast to the place where the skills and knowledge learned will eventually be applied?
A clear example of this appears in Piya Sorcar’s “Teach AIDS” work; she uses cartoon characters, instead of video or still images of real people, to depict patients and doctors having difficult discussions about AIDS, conversations learners are supposed to mimic.
The cartoon characters, and somewhat exaggerated voices, are less life-like than video-taped actors, but this brings the project two main affordances: people uncomfortable with the topic will probably be less uncomfortable watching cartoons, and the cartoons can be more readily reappropriated for other cultures, languages, etc. In fact, as Scott McCloud notes in his analytical work Understanding Comics, viewers are more likely to bond with a less specific representation than a nuanced, detailed representation, and this explains much of the popularity of cartoons, comics, avatars, anime, etc. Without surface details that allows us to categorize a person as a specific type, a simplified cartoon allows us to empathize and not get focused on minute facial details. So perhaps a distinction should be made between the form that learning media takes and the form that a learning practice or simulation takes. For example, in a blended curriculum using Teach AIDS, a teacher may want students to practice the conversational skills they’ve learned from cartoons in a highly realistic environment with a doctor’s office. Or, perhaps realism should scaffolded; students start enacting a highly structured, theatrical dialogue with cartoonish props, etc. and eventually make their way to a realistic imitation.
This also would seem to be the case with narrative; narratives simplify life and often provide causality to situations that are in fact much messier. Real life rarely structures itself into three acts with a climax and resolution. But by learning narrative, we are given a lens onto the world that structures events and lets us understand interpersonal dynamics, for example. “Oh, the tension is rising between these two people; what will be the climax of their conflict?” So it seems that the learning actually shapes our perception, and our experience of “real life.” I’m curious how much the theory of encoding specificity grapples with these complications.
