Saturday, June 11, 2005

Minds and algorithms

When minds are compared to computers, it is intuitively done by assuming that brains have much higher parallel processing power then computers have (indeed computers work highly sequential). But I must come back asking myself where does the power of minds come from ? How does that parallel processing work? Why does languages seem to hide this huge activity behind the scenes ? When we use words like, "me" or "room", we don't feel any multiple activity at all. We just know what we mean, we know what kinds of objects to take from our memories. I think one reason why scientific thinking has been so succesfull was to reverse that process: First trying to understand a proccess and then try to 'label' it. Maybe this is also why science education is difficult to understand for some children, as they use 'labels' for processes which they don't yet understand.

Another key idea in the theories of Minsky is the working of multiple representations and the switching among them. I also like that idea that the aim of education should mainly focus on learning to build more robust networks of representations and learn new ways to build those representations.