To imagine or Not to Imagine? A Meta-Analysis Investigating the Effectiveness of Mental Simulation on Behavior

Mental simulation is an important tool for managers who want consumers to imagine what life would be like if they engage in a new behavior. However, academic research found inconsistent effects of mental simulation on behavior. To understand this inconsistency, we conduct a meta-analysis and quantify the effect of different mental simulation prompts on behavior. Based on a cross-disciplinary dataset of 166 effect sizes spanning four decades (1980-2020) and representing 23,488 respondents, we identify three distinct simulation paradigms, quantify their effectiveness, and map them onto marketing practice: Studies that prompt simulation once (most common in academic research and akin to A/B tests in practice) revealed a very small, positive effect on behavior. Studies that used repeated and spaced prompts (e.g., weekly, akin to real-world marketing campaigns) yielded a large-sized positive effect. Studies that used repeated and massed prompts (e.g., in the same session, akin to encountering high-frequency ads online) revealed a medium-sized but negative effect on behavior. Further, imagery especially increased behavior in studies with enhanced visual prompts (e.g., AR) and studies that used moderately familiar targets, material purchases, or indulgent food. We explain the implications of these findings for theory and practice and identify novel avenues for research.