We demonstrate that it is relatively easy in a real life situation to make reasonably intelligent adults believe that they have witnessed something they actually have not seen themselves, but only heard reports about from others, and to make them report about particular details of the event. The event concerns the crashing of an El Al Boeing 747 on apartment buildings in Amsterdam. Over sixty per cent of the subjects said they had seen the crash on television, although no television film exists. Unexpectedly, women proved themselves significantly more vulnerable to this effect than men.
Reports by eyewitnesses are among the most important types of evidence in criminal as well as in civil law cases. Many cases are decided exclusively on the basis of eyewitness testimony. It is therefore disturbing that such testimony is often inaccurate or even entirely wrong. Errors may originate at the moment of perception and storage, during retention, and/or at the time of retrieval. At the moment of perception the witness may not see or hear everything there is to see or hear, or, alternatively, fail to store significant parts in long-term memory. The difference between these two theoretically distinguishable possibilities, however, is moot as we are not able to distinguish between them empirically.
A similar problem occurs when we try to distinguish empirically between three theoretically possible sources of errors during retention. During retention, (a) a witness may simply forget what was perceived, (b) information from another source-post-event information-may replace part of the already stored information, or (c) be added to it. The fact that forgetting does occur and that retention time is an important-but by no means the only-factor influencing forgetting, hardly needs documentation (Baddeley, 1990;Cohen, 1989). Memory distortion through post-event information is also a well-established fact since Elizabeth Loftus (1979) did her well-known experiments on the topic.