Eyewitnesses were badly wrong about the shooting of Jean Charles de Menezes
If someone was killed in front of you would you remember what happened? Many experts are challenging the view that eyewitnesses recounting what they saw is the best way of tapping their memory. Some think brain scans could be the way forward.
Think of a journey you made yesterday. I'm sure you remember it.
So can you remember whom you sat next to? Can you remember what the weather was like? Who was in front of you in the petrol queue? Was it a man or a woman?
FIND OUT MORE...
Law in Action is on Radio 4 on Tuesday, 1600 BST
Naturally, most of the time we don't remember these details. But what if someone got knifed in the petrol station? Then we become witnesses to a crime. And our ability to recall these minor details may have a significant role in authenticating our memory of the offence.
Some researchers suggest that we shouldn't need to remember these details. They are increasingly questioning the way that the police, lawyers and the courts think about memory. They argue that this conventional model of memory – like a detailed photograph or video film – is fundamentally flawed.
One of the most prominent of these researchers, Prof Elizabeth Loftus of the University of California at Irvine, even says that courts should have a new oath for witnesses: "Do you swear to tell the truth, the whole truth, or whatever it is you think you remember?"
Now Prof Martin Conway, a cognitive psychologist at Leeds University, has drawn up a report for the British Psychological Society and the Law Society calling for a major rethink of memory and the law.
He suggests his guidelines will help scientists who specialise in memory research when they testify as expert witnesses to help the courts assess the evidence.
Neuroimaging could offer a way forward in court
Memories are essentially a construct from a variety of sources and experiences, Prof Conway says. They are not necessarily a factual account of what happened.
What's more, a significant proportion of people seem to be highly suggestible and will quite readily change what they remember if given appropriate cues.
In one famous study, Dutch researchers questioned people about a 1992 accident in which a cargo plane had crashed into a block of flats near Schiphol Airport.
Ten months later, they conducted a survey asking if people remembered seeing the TV film of the plane hitting the building. More than half of the respondents said they had. A later study found that the proportion had gone up to two-thirds.
The problem is, there is no TV film of the accident. Asking the question had itself apparently changed people's memories.
A similar phenomenon happened with the shooting in London of the suspected terrorist Jean Charles de Menezes at Stockwell Underground Station.
Jean Charles de Menezes: Eyewitnesses claimed to have seen bulky jacket, victim jumping barriers - not correct
Jennifer Thompson: US rape victim identified wrong man with absolute certainty - suspect exonerated by DNA evidence
Initially witnesses claimed that he was wearing bulky clothing and that he had vaulted the ticket barriers as he ran from police.
A police spokesman said on the day that, "his clothing and behaviour added to their suspicions", and that he ran onto the train after police had issued warnings. These claims were incorrect.
But people still express surprise when told he wasn't wearing a large coat and are confused about how he entered the tube because the inaccurate reports became cemented into individual memories.
So are witnesses consciously or subconsciously having their memories altered?
There is little data available regarding the extent of suggestive questioning of eyewitnesses. One British study using actual interviews indicates that approximately one out of every six questions posed to eyewitnesses was in some way suggestive.
The police say they are already aware of the risks and do their utmost to avoid them.
At Hendon Police Training College, in north London, one of the courses trains officers in interviewing eye witnesses.
The officers are shown a video of a dramatised murder and then questioned by a colleague. One half-remembered a northern accent. When questioned further, she said she thought the accent was Mancunian but she couldn't be sure. The interviewer pressed the would-be witness on the detail despite the original uncertainty.
Our faculties are often impaired by the nature of the scene we are witnessing
Prof Conway argues this sort of detail can mislead and that the interviewing process could turn a doubtful memory into fact.
One technology that could help in future is brain scanning. Neuroimaging has now been developed in which objects unique to a crime scene are shown to witnesses, such as a lampshade or a particular colour.
These would only be recognised if the person had been there. Witnesses' brains are monitored to see if areas associated with memory light up when they see the objects. But it will be many years before such evidence is admissible in court.
And many involved in the criminal justice system argue that the courts already have a good record of separating reliable and unreliable memories.
Former judge Gerald Butler QC says jurors can use their commonsense to decipher the evidence and he is wary about introducing a new set of experts into the process.
"I have had a long term concern about the evidence of experts," he says.
"I have heard so many experts giving evidence one way and then another expert giving evidence the other way. It's very difficult to judge the expert and know which one is right and which one is wrong."
Send us your comments using the form below.
While working in an Afghan refugee camp in the 1990's, I unfortunately, witnessed a shooting of a man with an automatic rifle. We had a doctor in the vehicle and pulled up to the victims van to see if we could help. Some 30 rounds had hit into/around the victim and I was window to window with him, but from just after the event to this day I have NO visual image of the body, only a vivid memory of a bullet hole in the metal at the top of the windscreen. In this case my mind seems to have 'stored' that visual file. WHY?
Brett Tricker, uk
I heard a story once about Bono from U2 going to visit his wife after she had given birth and a nurse ran up to him pointing and shouting to the sister "This is him! This is the man who has been stealing all the bed pans." Once it had been pointed out that he could buy the hospital if he wanted the nurse accepted she may have got it wrong!
Any one that has read any of the good Doctor Robert Anton Wilson's books will already be familiar with the problem of memory recall and witness statements/ His experiment from the 1960s involved him on stage talking and a coloured man would rush on stage with a banana in hand and pretend to stab him, when the audience got over the shock and realised it was part of the talk, he would get them all to write down what happened. Inevitably there would be a hundred different statements with different stories usually revolving around a black man stabbing the speaker with a huge knife or even shooting him. To further show the unreliability of memory he would then get everyone to write down everything they saw from entering the theatre to sitting down, and again there would be hundreds of differing accounts of the colour of the theatre, the size and even the colour of the carpet. Memory/false memory and manipulation is a very easy function to manipulate.
hagbard celine, brighton
I used to be a litigation lawyer and I used to maintain that there were at least six versions of the truth: 1) what your client genuinely believed was the truth; 2) what the documents suggested to me was most likely to be the truth; 3) what the client on the other side genuinely believed was the truth; 4) what the documents suggested to the lawyer on the other side was most likely to be the truth; 5) what the judge would decide the truth had been on the basis of everything before him or her; and, lastly and almost least importantly, 6) what the truth actually was!
Ronnie MB, West Norfolk, UK
I was knifed when I was a teenager. The police asked me to describe the man, which I could (down to telling them, correctly, that he'd had a haircut since the photo I picked out in the photo ID). However, I thought he was a couple of inches taller than me, when actually he was a couple of inches shorter - I was told later that fear changes your memory, making him seem bigger.
Rona Moody, Pittsburgh, USA
I don't think it's memory loss. Its the case of bias interpretation of what was seen. there are three things, what I saw, what I think I saw, and what do I want to tell you what I saw.
amer, London UK
I lost my memory of military operations, whist on active service 40 years ago. Recently, those memories returned with vigour and full detail. I recalled each action and each word spoken at that time. It was as thought events had occurred only yesterday. I think some memories are ingrained and hypnosis may be the key to reveal the true facts of events.
Henry MacGowan, Glasgow, Scotland
There is a fair bit of research which confirms how inaccurate most people's memories are, although I can well believe that little of it relates to suggestive questioning by police and other authorities. As the article notes, people are also very suggestible and like to agree with questioners. Leading questions produce wrong answers. As well as misremembering what they saw or didn't see, interviewees also get wrong the dates of key events by wrongly referencing them to other events: "It must have been in 19## because my son was born that year..." and the like.
John K, Exeter
Aren't suspects called as witnesses? Does that mean that in the future suspects might have to be told that they have the right to remain silent AND the right not to undergo a brain scan while being shown images related to the crime? Otherwise (assuming this technique is reliable), they would be incriminating themselves.
As Grissom in CSI says, witnesses lie. They get confused. They're prejudiced, whether consciously or unconsciously, and they make mistakes. Eye witness evidence is unreliable, and it's cold hard science that should be relied on in court. That's why clear CCTV evidence is so important.
The BBC may edit your comments and not all emails will be published. Your comments may be published on any BBC media worldwide.