Memory for order is fundamental in everyday cognition, supporting basic processes like causal inference. However, theories of order memory are narrower, if anything, than theories of memory generally. The memory-in-chains (MIC) model improves on existing theories by explaining a family of order memory effects, by explaining more processes, and by making strong predictions. This paper examines the MIC model's explanation of primacy and recency effects, and the prediction that primacy should

more »

... ate recency. This prediction is supported by existing data sets, suggesting that Estes's (1997) perturbation model, dominant among theories of order memory, is incorrect. Fits to data are presented and compared with fits of other models.