Narrative characteristics of autobiographical memory predict expectations of narrator gender

Gender differences in autobiographical memory have been documented in various domains. The current study attempts to sharpen an understanding of extant gender differences in autobiographical memory by considering listener expectations, contributing to a sociocultural model. Across three studies (a pilot and two separate data collections) a narrative database was created. Narratives were coded for variables that have been shown to differ by gender, including emotion, connectedness to others, factual and interpretive elaboration, and thematic coherence. Participants read narratives and assessed if the narrator was male or female, indicating their confidence. In the pilot study and study 1, narrative features consistently predicted participant's assessments of narrator gender in two separate methods using the same narratives. In study 2, new narratives were used, showing replicability of the method and avoiding potential confounds of naturally occurring differences in study 1 and pilot data. More emotion words predicted higher likelihood of assessing a female narrator. Results support the use of such a methodology as an innovative way to examine gender-based listener expectations in autobiographical memory. They can be used for further theorising about the source of gender differences in this domain of memory that represents a combination of cognitive and social processes.