The process-dissociation procedure and similarity: Defining and estimating recollection and familiarity in recognition memory

In L. L. Jacoby's (1991) 2-study-list process-dissociation (PD) procedure, recollection (R) is estimated by the difference in responding between including and excluding items from previously studied lists. Familiarity (F) is estimated with the estimate of R and equations representing the assumption that F and R are independent. The authors hypothesized that ability to exclude items from 1 list (and hence R and F) depends on similarity between lists. As predicted, R was greater and F was smaller when the 2 study lists were encoded with different tasks rather than the same task. Thus, R in the 2-study-list PD procedure does not necessarily involve retrieval of all attributes of the study episode, and F is not necessarily a completely undifferentiated feeling of oldness. Dividing attention at test (Experiment 2) or study (Experiment 3) dissociated F and R. Results are interpreted using a functionalist approach to recognition memory that combines aspects of dual-process and global memory models.