Cognitive Effects and Correlates of Reading Fiction:Two Pre-Registered Multi-Level Meta-Analyses

Despite the ubiquity of reading fiction, the psychological effects of this mental activity have been debated. We present two pre-registered meta-analyses that synthesise cognitive effects and correlates of reading fiction. In meta-analysis 1 (368 effect sizes/69 experiments), reading fiction led to significant small-sized cognitive benefits, g = 0.14, 95% CI [0.07, 0.21], p < .001. Comparison group (watching fiction/reading nothing > reading nonfiction), outcome variable (significant positive effects emerged for empathy and mentalising only), and publication status (published > unpublished) were significant moderators. In meta-analysis 2 (551 effect sizes/111 correlational studies), lifetime exposure to print fiction was linked with significant smallsized cognitive benefits, r = .17, 95% CI [.13, .20], p < .001. Outcome variable (verbal abilities > general cognitive abilities > remaining outcomes), fictionality of the print material (fiction > nonfiction), publication status (published > unpublished), risk of bias (low > medium risk of bias), participant group (community > student samples), fiction exposure measure (Author Recognition Tests (ARTs) by Acheson et al. (2008) and Brysbaert et al. (2020)/self-report > ART by Grolig et al. (2020); ART by Acheson et al. (2008)/self-report > ART-Genres by Mar & Rain (2015)), study design (correlational > experimental), and percentage of female participants (via a negative relationship with cognitive benefits) significantly moderated the effect. Together, this set of metaanalyses provides robust evidence for a small-sized positive relationship between reading fiction and cognitive benefits.