For the experiment, Kroes took 39 patients who were undergoing ECT for severe depression (ECT works by passing electrical currents to parts of the brain, which triggers a brief seizure; patients are given muscle relaxants and an anesthetic). Each patient was asked to watch two rather upsetting videos, one about a child who is hit by a car and has to have his feet severed by surgeons, the other involving a pair of sisters, one of whom is kidnapped and sexually assaulted.
A week later, the patients were asked to recall details about one of the two stories (not both), after which time they were randomly sorted into three groups, A, B, and C (a control group).
Members of Group A and B were treated with ECT immediately following the retelling of the story. The following day, Group A participants had to complete a multiple-choice quiz about both stories. Fascinatingly, they did a better job recalling the story for which their memories hadn't been reactivated (or reconsolidated).
Their recall ability for the recounted story was no better than chance. Group B, on the other hand, had their memories tested about 90 minutes after the ECT, and their recall abilities were intact -- suggesting that it takes time to impair a memory. Group C did not receive ECT at all, and their recall abilities were solid, indicating that both ECT and reconsolidation is required to impair memory recall.