Masters Thesis

University of Missouri - Columbia, Department of Computer Engineering - Computer Science, 16 May 2003
Advisor: Ron Sun

Title: Simulating the interaction of explicit and implicit learning using an integrated cognitive model

Abstract: Since the 1980s, experimental psychologists, using a number of cognitive and motor skill learning tasks, demonstrated a contrast between improved task performance and the lack of corresponding explicit knowledge in human learning. Experiments have confirmed that in the situation when no apriori knowledge about the nature of the task is available, human subjects implicitly learn to improve performance and only develop explicit knowledge afterwards. One explanation for this phenomenon could be the existence of two separate structures in the human mind: one implicit, distributed, and not consciously available, and another explicit, crisp, and available for conscious retrieval. In the field of cognitive modeling, computer-assisted models are developed to simulate human learning, test hypotheses, and gain insight into cognitive processes. CLARION is a model based on the assumption of dichotomy of implicit and explicit learning, which uses a Q-learning neural network to represent the implicit learning structures, and a rule system to represent the mind's explicit part. This work presents CLARION simulations of a number of human process control task experiments, and examines the model's viability for simulating human learning when no apriori knowledge is available.

Full text in PDF

Supporting materials:

sw7d/pawel/academic/masters_thesis.txt · Last modified: 2009/11/21 22:23 by p77
CC Attribution-Noncommercial-Share Alike 3.0 Unported Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0