Eyewitness Memory

Event Testimony

One domain of particularly important work, pioneered by psychologist Elizabeth Loftus in the mid-1970s, concerns postevent information. Postevent information is “information” (either true or false) that an eyewitness might acquire after the event that is incorporated into later testimony as though it was part of what the witness had originally seen. One common example of postevent information is a misleading question. For instance, after seeing a man walk in through a front door and shoot a store clerk, an eyewitness might be asked if the gunman who came in through the back door said anything before he shot the clerk. When later asked to describe the actions of the gunman, the witness is more likely to say that he came in through the back door than if that question had not been asked. The phenomenon of people incorporating postevent information into their recollections of earlier events contrasts sharply with the legal system's general presumption that although a person might forget details—and therefore would not report them later—their memory would not incorporate new details.

There remains some debate among psychologists regarding exactly how postevent information influences memory. The original memory could be actually replaced by the postevent detail, or it could be blended with the postevent detail, or an entirely new, separate memory could compete with the original memory for retrieval. There are experimental results consistent with each of these processes, and it is possible that any of the three can happen, depending on the circumstances. In any case, there are clear regularities to the conditions that increase and [Page 341]decrease the magnitude of postevent information effects. For instance, postevent information effects are more pronounced when the original memory is weak rather than strong. Hence, it tends to be easier to find postevent information influences on peripheral details of an event than on central details of an event. Likewise, postevent information effects tend to be stronger when the witnessed event was in the more distant past. Furthermore, postevent information is more likely to be incorporated into memory testimony when the postevent information is plausible rather than implausible.

...