Examining the Effects of Emotional Valence and Arousal on Source Memory: A Meta-Analysis of Behavioral Evidence

The current meta-analysis examined the effects of valence and arousal on source memory accuracy, including the identification of variables that moderate the magnitude and direction of those effects. Fifty-three studies, comprising 85 individual experiments (N = 3,040 participants), were selected. Three separate analyses focusing on valence effects (valence-based: negative-neutral; positive-neutral; negative-positive) and other three focusing exclusively on arousal (arousal-based: high-low; medium-low; high-medium) were considered. Effect sizes varied from very small to medium. For the valence-based analyses, source memory accuracy was impaired for emotional compared with neutral stimuli (d(unb) = -.14 for negative-neutral; d(unb) = -.11 for positive-neutral), with a similar performance found for the negative-positive comparison (d(unb) = -.04). In the case of arousal-based analyses, source memory was improved for stimuli with high and medium arousal versus low arousal (d(unb) = .27, d(unb) = .49, respectively), with no statistically significant difference between high and medium arousal stimuli (d(unb) = -.12). Emotion effects on source memory were modulated by methodological factors. These factors may account for the variety findings typically found in emotion-related source memory research and could be systematically addressed in future studies.