Significance Testing of Rules in Rule-Based Models of Human Problem Solving

Rule-based models of human problem solving have typically not been tested for statistical significance. Three methods of testing rules ¿analysis of variance, randomization, and contingency tables¿are presented. Advantages and disadvantages of the methods are also described.

[1]  William B. Rouse,et al.  A fuzzy rule-based model of human problem solving , 1982, IEEE Transactions on Systems, Man, and Cybernetics.

[2]  Ronald A. Christensen,et al.  Contrivedness: The boundary between pattern recognition and numerology , 1982, Pattern Recognit..

[3]  William B. Rouse,et al.  The effects of type of knowledge upon human problem solving in a process control task , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  B. Margolin,et al.  An Analysis of Variance for Categorical Data , 1971 .

[5]  Sandra H. Rouse,et al.  A Rule-Based Model of Human Problem Solving Performance in Fault Diagnosis Tasks , 1980, IEEE Transactions on Systems, Man, and Cybernetics.

[6]  Allen Newell,et al.  Computer text-editing: An information-processing analysis of a routine cognitive skill , 1980, Cognitive Psychology.

[7]  William B. Rouse,et al.  A rule-based model of human problem-solving behavior in dynamic environments , 1985, IEEE Transactions on Systems, Man, and Cybernetics.