Memory is a composite of retrieved actual perceptions (“I felt the rain, I heard tires squeal, I saw a Volkswagen under a big red truck”) and subconscious reconstruction (or invention) if events not actually perceived, so as to form a coherent logical whole: (“This Volkswagen skidded on the wet street and crashed under a big red truck.”)
Often what an eyewitness “remembers” is largely a product of how a question is asked. (Compare “How fast was the Volkswagen going when it smashed into the truck?” versus “What was the speed of the Volkswagen when the two vehicles met?”)
The eye is not an instrument like the mirror… it is always self-conscious, always conditioned by some purpose–it is a sieve that allows many things to filter through in the process of seeing. The eye looks, it searches, and whatever it does not understand it does not see. A hunter does not see like a sailor. And somebody who never hunts may not see a rabbit even if it is sitting right next to him…
The accuracy of an eyewitness’ recollection is suspect when: there was little time for observation; the witness was under the influence of alcohol or drugs; the witness was in a highly emotional state; the witness would have had no reason to take particular notice of the incident (e.g. did not know a crime was taking place); there was a substantial time lapse between observation and first identification; there was a marked discrepancy between the identifying witness’ original description and the actual appearance of the defendant; the witness initially expressed doubt he or she could make an identification, or first made a patently erroneous one, or reported other details that later proved incorrect; there is steady “improvement” of memory and confidence from first identification through subsequent statements, grand jury testimony, and trial; the witness fails to recognize the defendant when next encountered; the witness and defendant are of different ethnic groups/races; the witness demonstrates bias or prejudice; the witness has an emotional need to see what he expects to see or report; and/or the witness manifests a high need to conform to, defer to, or please authority figures.
Standard mug-shot and lineup procedures unfortunately provide good illustration of how memory is commonly distorted. Typically, photographs (or lineup participants) shown to a prospective witness are all of persons suspected of the same crime as the offense under investigation, rather than persons fitting the descriptions originally given by witnesses. Thus, there tends to be a wide discrepancy in appearance among the suspects viewed, heightening the probability that the witness will identify the one face (relative to the others) that at least somewhat resembles the witness’ original perceptions. The police frequently use the photograph of the one they believe to be the perpetrator again and again in different displays, whereas the other faces reappear far less often; repetition alone can increase the witness’ conviction that the suspect’s face is familiar.
If an eyewitness erroneously chooses a mug shot after the initial lineup viewing, the witness is likely to choose that very same incorrect face once again at a later time, even though the correct face is also available for the choosing. That is, a witness is much more likely to identify in a real lineup that face he or she may have previously seen at the time of the initial identification, and in a subsequent recall is likely to remain committed to his or her error.