The Temporal Effects of Story and Statistical Evidence on Belief Change

Abstract

An experiment investigated characteristics of evidence that influenced beliefs across multiple time periods. Respondents, after reading one of four written messages that crossed evidence type (story or statistical) with vividness (vivid or nonvivid), completed belief measures at one of three time intervals (immediate, 48-hour delay, or 1-week delay). Statistical evidence was found to be more persuasive than story evidence. Vivid and nonvivid statistical evidence were persuasive relative to the control at 48 hours, and vivid statistical evidence remained persuasive through 1 week. Neither form of story evidence was persuasive relative to the control at delayed time intervals.