A Discrimination-Association Model for decomposing component processes of the Implicit Association Test

A formal model is proposed, which decomposes the Implicit Association Test (IAT) effect into three process components: stimuli discrimination, automatic association, and termination criterion. Both response accuracy and reaction time are considered. Four independent and parallel Poisson processes, one for each of the four label categories of the IAT, are assumed. The model parameters are the rate at which information accrues on the counter of each process, and the amount of information that is needed before a response is given. This study aims to present the model and an illustrative application in which process components of a Coca-Pepsi IAT are decomposed.