The impact of anecdotal information on medical decision-making

AbstractIn prior research, arguments using both anecdotal and statistical evidence are more persuasive than arguments using either alone (Allen, Bruflat, Fucilla, Kramer, McKellips, Ryan, & Spiegelhoff, 2000; Hornikx, 2005). However, it is less clear how people integrate information when the statistics and the anecdotes present conflicting information. In three preregistered experiments, we tested how people integrate conflicting information to judge the efficacy of a medicine in a clinical trial. Participants read either an anecdote from someone in the trial, summary statistics about the trial, or both types of information. We found that reading an anecdote from a member of the trial for whom treatment was ineffective reduced people’s beliefs in a medical treatment even when participants received strong evidence that the treatment was effective. In Experiment 3, we found that introducing icon arrays increased the perceived efficacy of the treatment but did not eliminate the effect of the anecdote.


Return to previous page