How much trait variance is captured in autobiographical memory ratings?

Abstract

This study examined the trait-variance in autobiographical memory recollection (AMR). Participants (N = 397) provided ratings on eight autobiographical memories elicited using cue words. Multilevel random coefficients modeling revealed that 43% of the variance in memory ratings was due to individual differences, while the remaining 57% was related to other factors. Scores on the Brief Autobiographical Recollection Test (Brief ART; Berntsen et al.), a measure of metacognitive evaluations of autobiographical memory, were significantly correlated with mean memory ratings. However, they only explained 34% of the variance in mean ratings, and 15% of the variance in the ratings of individual memories, suggesting that metacognitive evaluations do not fully coincide with actual memory experiences. These findings suggest that AMR is a complex phenomenon influenced by both stable individual differences and situational factors. Multilevel modeling provides a rigorous approach to gaining a more nuanced understanding of AMR by disentangling various sources of variance.