For the computer nerds in the house: A perceptron!

We came across a fascinating article in the New York Times about perceptual learning.

The upshot: researchers and teachers are using rapid-paced pattern recognition tasks to help students learn abstract concepts. For example, instead of doing traditional, paper-based math problems in order to learn the relationship between an equation and a graph (the way I learned algebra in high school), the student uses a computer program that presents an equation and a set of graphs. The student has to choose the graph that matches the equation — and the student has to go fast, moving through a series of such problems.

The idea is to develop a “gut instinct” – to recognize patterns – in order to scaffold a student’s ability to understand the abstractions and principles being taught. It’s a fascinating “bottom-up” approach to teaching “top down” ideas.

When I was in school, we learned top down ideas, well, top down.

I am not surprised these techniques work. I’m no expert, but I did study artificial intelligence and cognitive science, and I did some neuroscience research after college. A key lesson for me: the brain is really good at sifting through and identifying patterns, even in really noisy, messy environments. When this pattern recognition faculty is joined up with abstract reasoning, pretty awesome things can happen. Indeed, when these two styles of thinking come together in the right way, it feels fun.

Games like chess and poker bring all three of these things together (pattern recognition, reasoning and fun), and we all correctly admire and respect the true masters of these pursuits. I’d say we accord this respect because we think that mastery of these games points at skills we think we all ought to have, no matter what we do in life.

Looping back to perceptual learning: At Sifteo, our goal is to provide fun and engaging experiences that exercise different thinking skills. We hope to explore how we can apply perceptual learning ideas to our games – after all, our combination of interactivity, tactility (and fun) seems really well suited to the basic tenets of perceptual learning.