Assessing automaticity

We propose two principles that should be followed in the study of automaticity for cognitive processes. Both follow from the general rule that experimental research should be guided by a model of the task in question, frequently a process model. The first is that the concept of automaticity is best applied to component processes of complex behaviors rather than to behaviors as a whole. The second is that the criteria chosen for the identification of automaticity should be motivated by the processes in question. Examples are discussed of research programs that are relevant to each principle.