Author's School

Graduate School of Arts & Sciences

Author's Department/Program



English (en)

Date of Award


Degree Type


Degree Name

Master of Arts (MA)

Chair and Committee

Henry Roediger


In the categorized list procedure, subjects study words from semantic categories, then take a recognition test on those items. Subjects are likely to recognize common: high output dominance) category members even when they are not studied. Across three experiments, we sought to extend the categorized list procedure, further develop an explanation of why false recognition of category members occurs in this procedure, and modulate false recognition of category members by manipulating encoding and retrieval phases. Experiment 1 extended previous categorized list research, showing that subjects are likely to false alarm with high confidence to high output dominance category members that were not studied. Experiment 2 attempted to improve subjects’ metacognitive awareness of the deceptive nature of categorized lists by eliminating unrelated lures from the test list. This manipulation decreased the overall false alarm rate, but did not eliminate the relationship between output dominance and false alarm rate. Experiment 3 sought to reduce false recognition by providing subjects additional study and test events on the material. Providing subjects feedback on their tests followed by an additional study period to relearn the material was generally successful in eliminating high confidence false alarms at final test for frequent category members that were not studied. The results converge to imply that: 1) the categorized list procedure is a robust method for investigating false recognition,: 2) false recognition for high output dominance items may be related to source monitoring errors during testing caused by processes at encoding: e.g., Dewhurst, 2001), and: 3) these high confidence false memories are relatively resistant to manipulations intended to minimize them.


Permanent URL: