Learning Objects Aren't Legos, Part II

In my last post, I agreed with Stephen Downes that we have to be careful not to take our analogies too literally and specifically pointed out flaws in the “learning-object-as-software-object” analogy. Sometimes the best way to make sure an analogy doesn’t get too deeply rooted is to counter it with another analogy that causes just enough cognitive dissonance to keep you from getting intellectually lazy. (The classic example of this is the light-as-a-particle/light-as-a-wave analogy pair.) I’m going to attempt to create that kind of duality by balancing my course-as-a-sentence-of-natural-language analogy with another one.Composer and music historian Robert Kapilow uses a wonderful analogy to explain the course of music history (and art history, and literature history). He talks about something he calls “the genre moment.” Basically any genre of any art form has certain distinct characteristics that define genre for that historical moment. For example, there is a moment in many murder mystery novels when the suspects are all gathered in a room and the detective is about to reveal who dunnit. Now, most hacks will work strictly within the expectations of the genre moment, e.g., you will inevitably find out that the butler did it. But a real artist like, say, Agatha Christie, will play with our expectations regarding the genre moment. For example, we might expect that the butler did it but be surprised to find out that, actually, it was the daughter-in-law. Soon, lots of lesser artistic lights immitate the innovation of the leader, e.g., we suddenly see lots of mystery novels being written in which the ending is a genuine surprise. That becomes the new genre moment–the moment at which we have to guess which person is about to be revealed as the murderer.

Kapilow employs this analogy to great effect in a schtick he calls “What Makes It Great?”. He is a master at distilling the art in a musical composition down to a few chords or notes. “Anyone else would have written this,” he says impishly in his squeaky, nasal voice. “But Mozart wrote this.” By changing one little detail in the composition, Kapilow suddenly makes the magic of the musical moment clear and concrete. In effect, he tells us, “Please pay attention to the man behind the curtain.”

We can think of learning experience objects as genre moments. There’s the moment when the teacher proposes a mystery to be solved by the class. There’s the moment when students negotiate a consensus answer in small groups. There’s the moment when students demonstrate command of the subject-matter through a test. We string these moments together into a course experience. A course in cellular biology may use some different learning experience genre moments than a course on Shakespeare (e.g., a lab), but there is still a finite number of experience objects that can be combined generatively to create a language of teaching and learning experiences.

Presumably, these learning genre moments are linked in some direct and/or indirect ways to the ways in which our brains need to sequence information and experience in order to learn. In other words, some decisions about sequencing have to do with the ways in which we are biologically hard-wired to learn, but the link can be mediated by fairly complex cultural issues of the moment. But we don’t have to fully understand these deeper links in order to to take advantage. We can gain advantages in both learning effectiveness and re-use in creation of online learning tools simply by observing the surface structure of the course experience sequences (to mix metaphors), assuming only that, after millions of years of teaching and learning, human beings have some tacit knowledge of what works and what doesn’t. To mix metaphors, deducing “grammatical” sequences of learning experience objects can start as an exercise in anthropology and may or may not evolve into cognitive psychology later on.

Furthermore, just because we don’t have legos doesn’t mean that we can’t define and isolate re-usable structures. I once asked Rob Kapilow whether he first “hears” the emotional impact of a composition and then looks at the compositional structure or the other way around. His reply was, “I don’t understand the difference.” Now, Rob was Yale University’s youngest ever assistant professor; there is very little that he doesn’t understand. What he meant was that the emotional impact that we experience from a work of art is simply the product of way in which our emotional brains parse the structure of the composition. We experience vibration at a certain frequency as sound and we experience a series of notes and chords as a moving piece of music.

Likewise, just because our brains are tuned to experience a sequence of learning moments organically and holistically doesn’t mean that the structures of the atomic experiences can’t be specified and isolated. In fact, what it indicates is that the structures we are attempting to identify are represented in some real way in our unconscious brains. This is exactly how, for example, linguistics defines itself as a discipline. Linguistics is the study of what it is that we know when we know a language. It’s an attempt to make deeply and fundamentally tacit knowledge explicit.

I do think we have a better shot, in the long run, of developing a sort of linguistics of learning for learning experience objects than for learning content objects. The reason goes back to my analogy in part one to content as semantic features and experience as syntactic features. We have some reason to believe that syntax and semantics are represented quite differently in the brain. Semantic functionality is mostly not localizeable within brain anatomy, and the reason may well be because meaning is represented as a web of weighted connections. This sort of representation, known in cognitive theory as connectionism, seems to fit well with the whole problem of meaning shifting almost infinitely depending on context. However, syntax appears to be much more localizeable within brain anatomy and seems to have functional characteristics that are highly similar to the ways in which software programs run, right down to the presence of a parser.

My intuition is that the ways in which we can string together learning experiences are much more like like the ways in which we construct (and parse) sentences for grammaticality than the ways in which we extract full meaning. Think of syntax as a series of rules that give us cues for the kinds of words that can be meaningfully strung together. For example, a sentence must have a subject and a verb, and the subject must be a noun. Note that the content can be quite diverse and still fit the structure: “I eat,” “Sh*t happens” and “Creativity inspires” are all grammatical sentences with the same syntactic structure. The rules of syntax simply provide containers that enable us to string together blobs of meaning in ways that, together with context, makes it possible to constrain the range of possible meanings enough so that its even possible to guess what somebody might mean. Learning experiences structure content so that we know which features to pay attention to and have clues about some of the ways in which content blobs can be associated in our brains. I introduce, I reinforce, I test.

While content may seem on the surface like the easier way to go in terms of re-use, I think that’s mainly because some of the relationships based on meaning are much more readily available as explicit knowledge. In contrast, sequences of learning (or communication) experiences are, like language processing, deeply tacit. We have to observe what we do in order to figure out what we know. In the end, I think the more fruitful path to re-usability will be to create software that offers affordances to create explicit sequences of learning experiences. This is not to say that content can’t be re-used; of course it can. But content objects will rarely be re-usable without modification, while I believe that experience objects (i.e., learning exercises that are instantiated in types of classroom activities or afforded by software functionality) can be re-used without modification in an analogous (though not necessarily isomorphic) way to how the sentence structure “subject verb” can be re-used.