Introduction

John R. Anderson Christian Lebiere

Abstract

ACT-R consists of a theory of the nature of human knowledge, a theory of how this knowledge is deployed, and a theory of how that knowledge is acquired. As will be elaborated later, ACT-R assumes that there are two types of knowledge--declarative and procedural. Declarative knowledge corresponds to things we are aware we know and can usually describe to others. Examples of declarative knowledge include "George Washington was the first president of the United States" and "Three plus four is seven." Procedural knowledge is knowledge that we display in our behavior but we are not conscious of. Procedural knowledge basically specifies how to bring declarative knowledge to bear in solving problems.

Declarative knowledge in ACT-R is represented in terms of chunks (Miller, 1956; Servan-Schreiber, 1991) which are configurations of elements that encode various things that we know. Figure 1.1 is a graphical display of a chunk encoding the addition fact that 3+4=7. This chunk can also be represented textually:

   Fact3+4
      isa     ADDITION-FACT
      addend1 Three
      addend2 Four
      sum     Seven

In the above text, Fact3+4 is just an arbitrary name given to the chunk. It is followed by a series of slots and associated values. The first slot, isa, is special and gives the type of the chunk. The remaining slots (addend1, addend2, and sum) are associated with this addition-fact type and their values define the 3+4=7 fact. It serves essentially to interassociate the slot values of the chunk. The S_ji's in Figure 1.1 refer to associative strengths between slot values and chunks and the B_i refers to the base-level activation of the chunk. These quantities are relevant to calculating the activation levels of such chunks. These activation processes determine the availability of the knowledge as Chapter 3 will develop.

Production rules specify how to retrieve and use such declarative knowledge to solve problems. As an example of procedural knowledge the model below gives a production set for solving multicolumn addition problems. This production set consists of a set of productions where each production is a condition-action pair. The condition specifies what must be true for the production rule to apply and the action specifies a set of things to do if the production applies. The conditions test for the state of the current goal and chunks in declarative memory while the actions can change the goal state.

Addition Model