Individual differences in incorrect responding and the ability to discriminate the source of the products of retrieval.

Abstract

When memory is tested, researchers are often interested in the items that were correctly recalled or recognized, while ignoring or factoring out trials where one "recalls" or "recognizes" a nonstudied item. However, intrusions and false alarms are more than nuisance data and can provide key insights into the memory system. The present article reports 2 experiments demonstrating that people are remarkably consistent in the rate at which they incorrectly report memory for nonstudied items, even across a range of differing stimuli and test features. Experiment 1 found that individual differences in false alarms and intrusions were strongly related, even while controlling for the shared influence of memory ability on these incorrect response types. Furthermore, intrusion rate was found to be related to response bias as well, but this effect was suppressed by the shared influence of memory ability, demonstrating that an independent measure of memory ability is an important control in investigations into individual differences in response bias. Experiment 2 revealed that the relations between intrusions and false alarm rate and response bias were entirely explained by one's ability to discriminate episodic memories from semantic generation. This work links together previous work on individual differences in intrusions and in false alarms, and highlights the ability to identify the source of a memory as the key cognitive trait underlying incorrect response styles on various memory tests. (PsycINFO Database Record