|Symbol||Chapter 2||Chapter 2|
|Tutorial Units 1 & 2||Tutorial Units 1 & 2|
|Subsymbolic||Chapter 3||Chapter 3|
|Tutorial Units 4 & 5||Tutorial Unit 3|
|Symbol||Chapter 4 & 5||Chapter 4|
|Tutorial Unit 6 & 9||Tutorial Unit 8|
|Subsymbolic||Chapter 4||Chapter 4|
|Tutorial Unit 6||Tutorial Unit 7|
As can be seen from the table, this chapter on knowledge representation will also discuss the performance assumptions of ACT-R at the symbolic level. This is because assumptions about how knowledge is represented are not meaningful until they are tied to how the knowledge will be used. The symbolic performance assumptions in this chapter are not the only consequences of the ACT-R representational assumptions. They are just the most immediate consequences. These representational assumptions will also serve as the framework for the other ACT-R process assumptions.
This chapter will contain sections to discuss the procedural-declarative distinction which is the most fundamental assumption in the ACT-R theory, the representation of declarative knowledge as chunks, the representation of procedural knowledge as productions, and the critical role of goal structures in organizing behavior. At all points in the development of the ACT-R theory we like to maintain an intimate relationship with data. Therefore, the chapter will end with a discussion of the Tower of Hanoi task and ACT-R model for it. This is a task that depends on performance assumptions at the symbolic level.
Tower of Hanoi