Can a placebo for a psychotropic drug help participants resist the misinformation effect? To answer this question, we gave participants a mixture of baking soda and water and told half of them that the mixture was a cognition-enhancing drug called R273 and told the other half that it was an inactive version of the drug. Shortly thereafter, all participants took part in a three-stage misinformation experiment. Compared with participants who were told that they had taken the placebo, the participants who were told that they had taken the drug reported improved cognitive abilities and were less susceptible to the misinformation effect. We provide source-monitoring and mindfulness accounts of our findings.
Placebos can produce powerful effects, including improving people's subjective reports of cognitive performance. However, there is no objective evidence to support these reports: To date, people have shown no placeboinduced improvement in cognitive performance compared with a control group. For example, in the only published positive placebo and memory study, all participants received a placebo, but half were told it was a memoryenhancing drug and the other half were told it was a placebo (Kvavilashvili & Ellis, 1999). Participants who were told that they had taken the drug (told drug participants) thought that their memories were better than participants who were told they had taken a placebo (toldplacebo participants); however, both groups performed similarly at recall. Kvavilashvili and Ellis concluded that their findings fit with pharmacological research showing that memory performance is difficult to improve.
It is possible, however, that Kvavilashvili and Ellis's (1999) placebo did not improve memory because the recall task they used was not sensitive to the way in which placebos work. Kirsch and Lynn (1999) suggested that placebos cause people to set in motion a behavioral sequence that produces the effects they think placebos cause. In other words, people cause their own behavior, but they (mis)attribute it to the placebo. Thus, for a placebo to benefit memory, participants must be able to adjust some behavior to improve memory performance but attribute that behavior to the placebo. A recall task does not readily allow participants to make those behavioral adjustments. Therefore, we hypothesized that a memory task that did allow participants who had taken placebos to make behavioral adjustments would show memory benefits. One such task involves the effect of misleading suggestions on memory performance.
In a typical experiment, participants witness an event and then receive either nonmisleading or purposely misleading postevent information (PEI). Later, they are tested on what they can remember about the event. Results reliably show the misinformation effect: Participants more accurately recall items about which they received no misleading information (control items) than items about which they received misleading information (misled items) (Tousignant, Hall, & Loftus, 1986).
Many social and cognitive factors influence the misinformation effect. Socially, the status of the person delivering the misinformation affects the likelihood that the recipient will adopt the misleading suggestion (see, e.g., Vornik, Sharman, & Garry, 2003). Cognitively, one's ability to keep track of-and then detect discrepancies between-the event and the misleading narrative also affects one's susceptibility to being misled (Lindsay, 1993). Source monitoring is the process of remembering information and attributing it to the correct source (Johnson, Hashtroudi, & Lindsay, 1993); as source monitoring improves, people are less likely to be misled.
On the basis of this research, we might expect that if a placebo causes people to set in motion behaviors that result in better source monitoring, they should be better at detecting and discounting the misleading information that causes the misinformation effect. …