Thursday, September 24, 2009

Middleton is vain

Don't praise yourself too much, Dr. Middleton

Problem Solving

What are strategies?
How do we develop them (or how do they develop in us)?
Given no situation is exactly the same as a previous one, how can we choose an appropriate course of action?
How can we explain mistakes?

Problem solving is plagued with tough questions like these. But at its core, it is concerned with 4 basic issues outlined by Polya 60 years ago:

1. Understanding the Problem (problem recognition, problem construction, the role of memory)
2. Develop a Solution (accessing memories; instatiation of schemata; compilation of complex plans, goals, strategies, procedures);
3. Trying out a Solution (keeping track of problem solving as it occurs, repairing problem areas; Goal resolution);
4. Looking Back (examining the overall solution process, labeling it and its part for future retrieval).

We use many different strategies to do the same kinds of things. This is evolutionarily advantageous given that: In the wild, many escape routes (or routes to gain food, mates, etc) may be blocked, necessitating a different strategy); and redundancy allows for backup systems to be called up when primary systems are inoperative.

Schema Theory Introduction

The things to remember about schema theory is that schemata are:

1. evolved for remembering, not for storing information (this is not to say that storage is unimportant, just that storage doesn't have to be efficient if search and retrieval is);
2. frameworks that organize experiences into their general structural similarities, not carbon copies of experiences.
3. incomplete, needing local, immediate information to instantiate them, as well as some facts and rules retrieved from associative memory;
4. Hierarchical in that some schemata subsume others. A few good schema, when recombined can provide a kind of combinatoric power to aid memory and behavior;
5. Not the only way to think about memory. Schemata are models, not memory itself.

Friday, September 11, 2009

Introduction to production systems and schemata

Great class yesterday.

3 points to remember:

1: We place structure on our memories that help us get around its inherent limitations in capacity and processing speed: a) We classify similar situations and procedures into schemata; b) we chunk information to aid in complex behavior by overcoming limitations in STM; and c) the more specific our goals, the more narrow (and therefore less transferable) our memories become.
2: The metaphor of the mind as a computer is very powerful. Structure, in the form of production systems helps us explain how schemata are organized and interact. Condition-action pairs are the basic units by which logical operations are performed under this METAPHOR.
3: A reasonably small number of general schemata are sufficient (in their myriad combinations) to drive a large range of behavior. Some plans (or master programs, or superordinate production systems) "call" on other productions to compile a string of behaviors. The local information (like the size of the angle to be turned in our example) is placed in "slots" to "instantiate" these productions for action.

Wednesday, September 2, 2009

Welcome to Cognition and Instruction

This blog is meant for students in the doctoral seminar, DCI 791: Cognition and Instruction.

The first discussion we will be having is on just what a cognitive approach to education research and development is, and the kinds of questions asked by cognitive researchers. Please read the Greeno, Collins and Resnick article as well as Donald Norman.