Author(s), Title and Publication
Skurnik, I., Yoon, C., Park, D., & Schwarz, N. (2005). How Warnings about False Claims Become Recommendations. Journal of Consumer Research, 31(4), 713-724.
A large body of prior research from psychology and consumer behaviors suggested that exposing people to claims increased the perceived truth of the claim when it was seen again at a later time. This effect was found to be prominent even when the claim was explicitly identified as false in prior exposure. This illusion of truth effect can be explained by a constructive inference people make: when people have little information based on which to judge the truth of a claim, they often rely on partial information in memory, such as the claim’s familiarity. In other words, people tend to believe a claim to be true because they have seen it before.
Building upon this background, this research suggests that when people recall the truth of a claim, memory for the original context of the claim can be as important as memory for the claim itself. When people do not recall the original context or source of the claim, they tend to think the claim is true because of prior exposure, especially for older people. Through two experimental studies, this study documented a paradoxical effect of warnings: the more often older adults were told that a given claim was false, the more likely they were to remember it incorrectly as true after several days. Similarly, warning older adults that a previously seen claim of unknown validity is false increased acceptance of this claim as true, compared to warning them about false claims that they have not read before.
In the first experimental study, two groups of young (ages 18 – 25) and older (ages 71 – 86) adults were exposed to a randomized group of true and false statements about medicine and health. During the study, each statement appeared on the computer screen for 5 seconds, followed by a blank screen for 750 milliseconds (ms.), then the word “true” or “false” for 1,500 ms., and finally a blank screen for 1,500 ms. before the next claim appeared. Some statements were repeated three times while others appeared only once. Following the study, a memory test was administered 30 minutes (short delay) or three days (long delay) after the initial exposure. In a second study, a similar experiment was design except that 1) the truth or falsity of the statements was not disclosed until the last presentation, and 2) a memory test was administered only after a short delay.
Results of study 1 showed when a claim was repeatedly told to be false, repetition helped older adults accurately remember the claim as false after a short delay. But contradictorily, when their memory was tested three days later, the repetition backfired in that the more times older adults had been warned about a claim was false, the more likely they were to misremember it as true. No such tendency was found to misremember true information as false. Results of study 2 showed that when the truth or falsity of repeated statements was not disclosed until the last presentation, repetition led to poorer memory for both age groups. However, only older adults exhibited an illusion of truth bias.
Implications for Practice
With the prevalence of fake news and rumors, this study yields communication and public policy insights that are worth pondering. Attempts to correct people’s wrong beliefs about misleading information and rumors may have the unintended effect of increasing the familiarity of the false claim when it is mentioned again in a different context, rendering it more likely to be misremembered as true after the details have faded from memory, especially for older adults. Public relations professionals should communicate wisely and avoid statements that repeat the wrong claim, such as “It is not true that X is good for your arthritis,” or “Is it true X is good for your arthritis?” Alternatively, the recommended approach to counteract misperception is to reinforcement the accurate message, which will not suffer from this illusion of truth effect.
Location of Article
This article is available online at: