Learning Heuristically

Posted by on 30.09.2009 in art, creativity, education

Heuristic” — an ugly word that everyone should know — is used differently in various contexts.

It generally refers to a process of making open-ended, provisional decisions in order to get into a better position — “for now” — from which to act and decide better later:

Let’s see how this works out — find out if we learn something — and and then we’ll decide from there.

Think of it as navigating stepping stones to cross a river in the fog: we can’t always see exactly where we’ll be going two or three steps ahead until we make the step that’s right in front of us, then we assess our options from there, and then on to the next one and so-on.

Psychologists use the term “heuristic” in a more technical sense, referring to intuitive judgements and rules-of-thumb we use to make quick decisions. It’s also used in computer science to refer to algorithms that generate “good enough” solutions that aren’t perfectly accurate but save processing time.

The sense of “heuristic” that is most valuable here is taken from the early-20th century music theorist Henry Cowell. His idea was to learn how to play very complex rhythms by inventing a machine to play them first. Hearing these rhythms was the heuristic step before being able to play them.

If you’ve ever learned a complex skill like drumming, life drawing, dance, martial arts, etc, you will be familiar with the value of having models and demonstrations to learn from. We might have a vague sense of what we’re trying to do but can’t quite make it work until someone shows us how they did it; once we study the example it becomes so much simpler.

I’ve certainly experienced this with art. Without any instruction we draw people who look like sticks and blobs; but once we have access to a picture that someone else has drawn, we can learn by copying and mastering their ratios & relations and tricks of perspective.

Looking at the history of art we see how long it took to learn the rules for rendering three-dimensional objects, people, and landscapes in life-like ways. It required thousands of years of cultural learning.

Likewise, we can see a similar progression of ideas: geometry was a stepping-stone for calculus, which was a stepping-stone for Newton’s physics, which was a stepping-stone for industrialism, which was a stepping stone for so many features of the society we inhabit today — not the least of which being the hierarchically managed organizations through which basically everything is done.

Those organizations have worked well but technology has obviously progressed in ways that shatter a lot of their underlying structure. Things are certainly changing, and decisions we make going forward (i.e. which technologies to adopt and how) will effect how society continues to evolve.

Everything we do is a small vote that will determine, in the aggregate, the future of our society. In the near term, today’s decisions affect what options will be available tomorrow.

We can’t assume things will naturally progress and improve. Sometimes we step on the wrong stone and have to go back a few. For example, those artistic techniques that took thousands of years to learn were largely lost in the Middle Ages, and required several generations to re-learn during the Renaissance — even then with the help of classical models.

So where do we go from here? How do we know the most effective way to use the new platforms, practices, and ideas? In other words, what should we learn in order to learn better? How do we progress in order to progress more effectively?

Continued from The New Digital World View… to be continued.