Color association values and response interference on variants of the Stroop test

Abstract

A simple mediation model is applied to the Stroop phenomenon of response interference on color word naming tasks. A prediction is educed that as the degree of color association value for sets of words vary, the extent of response interference will vary directly. Five different lists, including three distinct levels of color associativity, were used in a repeated measurement Latin Square design. Results indicate a close correspondence between color association value and response interference. Word frequency is discussed as an important antecedent variable for response interference, whose role remains to be defined. It is suggested that a highly precise model for predicting interference times is feasible.