Two experiments attempted to resolve previous contradictory findings concerning developmental trends in false memories within the Deese-Roediger-McDermott (DRM) paradigm by using an improved methodology--constructing age-appropriate associative lists. The research also extended the DRM paradigm to preschoolers. Experiment 1 (N = 320) included children in three age groups (preschoolers of 3-4 years, second-graders of 7-8 years, and preadolescents of 11-12 years) and adults, and Experiment 2 (N = 64) examined preschoolers and preadolescents. Age-appropriate lists increased false recall. Although preschoolers had fewer false memories than the other age groups, they showed considerable levels of false recall when tested with age-appropriate materials. Results were discussed in terms of fuzzy-trace, source-monitoring, and activation frameworks.