Examining the impact of multiple tests on metamemory for emotional images

Abstract

Individuals engage in metamemory when they perceive, control, and monitor their memories. Metamemory research often focuses on participants’ accuracy in predicting future memory performance, commonly referred to as judgements of learning (JOLs). Participants often show higher JOLs for emotional (especially negative) content than for neutral content, but their recognition accuracy is often contradictory to these predictions, with better performance for neutral content. As JOLs may be the result of misunderstanding test conditions, participants may benefit from experience with test conditions to calibrate themselves better and adjust their JOLs to match their recognition accuracy. In the present study, participants studied a list of positive, negative, and neutral images while providing JOLs, and then completed an old/new recognition test. They then completed a second block of the same procedure, but with new images. JOLs were highest for negative emotional images in the first block, but recognition accuracy was highest for neutral images. JOLs were even less accurate on the second block, again showing the highest JOLs but the lowest recognition accuracy for negative images; experience did not calibrate participants to provide more accurate JOLs. Theoretical implications regarding the impact of emotion on metamemory for images and future directions of research are discussed.