Special events, such as the Kennedy assassination, are well-suited for comparative memory research. The day after the Challenger space shuttle disaster, students were asked to write down details about that specific day. Three years later, their memories were tested. Predictably, many students recalled various details incorrectly.
What was truly remarkable, however, was that several students were so confident in the accuracy of their memories from that day that they suggested their own written account from that time must be mistaken (Dietrich, 2007, p. 163).
Students watched a video of two cars colliding. They were then asked to estimate the speed of the cars, but the question was phrased slightly differently for each group. The wording varied: “How fast were the cars going when they hit each other, collided, smashed, crashed, or bumped?” The result: the more suggestive the verb used, implying higher speed, the higher the students’ speed estimates.
Even more surprising were the results of an interview with the same students a week later. When asked if there had been broken glass in the video, over half of the students exposed to the most suggestive speed-related question responded affirmatively—twice as many as those who had answered the least suggestive question. This incorrect memory seems to arise because recalling such events involves a reconstruction, with details that “fit” the scenario more likely to be added (Dietrich, 2007, pp. 162-163).