I will review the diffusion decision model (Ratcliff, 1978; Ratcliff & McKoon, 2008), which provides a detailed explanation of behavior in two-choice discrimination tasks. I will show how it translates behavioral data, namely accuracy and response time distributions for correct and error responses, into components of cognitive processing. Stimulus difficulty affects the quality of information on which a decision is based and instructions emphasizing either speed or accuracy affect the criterial amounts of information that are required before a response is initiated.
In both these cases, with only one parameter of the model changing, the model explains all the changes in accuracy and response time. One important feature of the model is that it fits data from a single 45-minute session with standard deviations in the main model parameters that are one third to one fifth smaller than individual differences.
This means individual differences such as relationships between model parameters and IQ can be studied. I will present applications of the model to aging, from children to 75-90 year olds, and show relationships between priming in item recognition and associative recognition with age and IQ. I also present results that explain why older subjects are slower in terms of the aspects of performance that are optimized.