THE UNIVERSITY OF SUSSEX CG019

BA & LLB FIRST YEAR EXAMINATION 2001

COGNITIVE MODELLING

12th September 2001 9.30 - 11.00

Candidates must attempt TWO questions
Each question is worth 25 marks.

1. 1. In the context of production systems explain the terms: working memory, production rule and conflict resolution. (7)
   2. The actions in the rules of Young and O'Shea's production system for subtraction that you studied were either procedure calls or tokens to be added to working memory. Explain the difference between these two kinds of action and indicate their implications for modelling children's subtraction performance. (8)
   3. Briefly discuss the strengths and weaknesses of Young and O'Shea's production system as a cognitive model of children's subtraction performance. (10)

2. 1. In the context of feed forward neural networks explain the terms: input layer, output layer, hidden unit, arc weight and node threshold. (10)
   2. A neural network that you have studied can compute the probabilities that a particular five letter sequence, such as gromo is either an English word or not. Explain how the system is trained to produce such a result. (10)
   3. What facets of the way neural networks process information have made them useful as cognitive models? (5)

3. 1. With the aid of examples explain the difference between a
      1. runnable and a non-runnable cognitive model. (4)
      2. symbolic and a non-symbolic cognitive model. (4)
      3. cognitive architecture and a problem solving model. (4)
   2. Explain the stages which the Blocks World language understanding program you have studied goes through in interpreting a question or command for the simulated robot to carry out. (13)