Adaptive Control of Thought-Rational (ACT-R):

 Adaptive Control of Thought-Rational:

Introduction

    ACT-R is a cognitive architecture mainly it is the most recent version of ACT. ACT-R is a model of information processing that integrates a network representation for Declarative Knowledge and a production system representation for Procedural Knowledge. This model was developed by John Robert Anderson in 1983.

Explanation

    In ACT-R, networks include "images of objects" and their corresponding "spatial configurations" and relationships. There is temporal information like relationships involving the sequence of actions, events, and order. Anderson called this information "Temporal Strings". 

The knowledge Within ACT-R 

Declarative Knowledge	Procedural Knowledge
Anderson’s Declarative Network Model contains a mechanism by which information can be retrieved.           It is a structure for storing information.	In Anderson’s Model Procedural Knowledge is represented in production systems rather than in semantic networks.
Nodes: In this model, the nodes can be either active or non-active. A node can be activated by an external stimulus like memories and thought processes.	Stages: There are 3 Stages for the representation of procedural skills.     Cognitive Stage     Associative Stage     Autonomous Stage
Spreading Activation: From one node to another in the network, there is spreading Activation in which each node gets stimulate from its neighboring nodes.	Proceduralization: It is an overall process by which we transform slow, explicit information about procedures into speedy and implicit implementations of procedures.
Ways of Network Changes: ACT-R also suggests ways by which the network changes as a result of activation.           Activation is likely to spread along the routes of frequently traveled connections.           It is less likely to spread along less used connections between nodes.	Production Turning: It is another aspect of proceduralization. It includes the two complementary processes of generalization discrimination. In this, one can learn to generalize existing rules to apply them to new conditions.