| A great deal of interest has focused on the use of Tulving's (1985) Remember-Know procedure as a means to study the subjective experience that accompanies memory retrieval. Theorists commonly assume that the experiential states of Remembering and Knowing measured by this procedure are direct manifestations of two underlying retrieval processes, recollection and familiarity. This assumption was put to empirical test. The recollective and familiarity processes were characterized by means of empirical findings and formal models taken from the wider memory literature. Based on this, predictions were made for the hit rate patterns of Remember and Know responses following associative and old-new recognition in three paradigrns: associative interference, part-set cueing, and retrieval practice. In Experiments 1--3, increasing the number of study-list items similar to a test probe (associative interference) decreased Remember hits and increased Know hits. In Experiment 4, cueing a test probe with similar items (part-set cueing) decreased Remember hits and increased Know hits. In Experiment 5, cued-recall of items belonging to a semantic category (retrieval practice) decreased Remember hits and marginally increased Know hits for unpracticed members of that category relative to baseline. These findings are consistent with the idea that Remember and Know states are directly related to the underlying retrieval processes of recollection and familiarity. |
