Ingrid Candel Ingrid Candel

How reliable is an eyewitness?

In Mind
Written by  Femke Kools Wednesday, 04 February 2004 00:00

Research into memory errors

Imagine: you stop for the traffic lights at a crossing. Right before your eyes a car hits a pedestrian. Directly after the accident you give a testimony as a witness at the scene. You describe the incident with all the details that are still very clear in your mind: the crossing pedestrian, the car driver who hit the pedestrian, the motorcycle with sidecar and the man who rode the motorcycle who gave first aid. Five weeks later you have to give your testimony again before the judge. You tell the same story, only this time your memory has already somewhat faded. You do describe the motorcyclist, but you forget to mention that it was a motorcycle with sidecar. To your astonishment your testimony is disregarded in the trial because it is a statement that cannot be used. The judge decides that it is inconsistent and therefore unreliable.

"Accurate eyewitness accounts are sometimes unjustly considered unreliable because they are inconsistent or incomplete," says dr. Ingrid Candel, researcher at the Faculty of Psychology of Universiteit Maastricht (UM). "However, accuracy, completeness and consistency are three independent variables. Although all three of them contribute to the reliability of a testimony, they do not exclude each other."

For years now, Ingrid Candel has been studying the functioning of the visual memory and the influence of that on the reliability of eyewitness accounts. In March 2003, she received a doctor’s degree for her research into errors that can occur in recollections of emotional events. In the beginning of January, the news got out that the Netherlands Organisation for Scientific Research (NWO) grants her a subsidy under the Innovation Impulse (Vernieuwingsimpuls) for the continuation of her research into memory errors. The NWO Innovation Impulse offers three types of subsidies: the veni subsidy for young, talented researchers, the vidi subsidy for experienced postdocs, and the vici subsidy for very experienced researchers. Candel will receive a veni subsidy that can finance her own position and that of a student assistant for three years.

In her thesis she defines the errors that can influence the three dimensions of reliability that were mentioned before. "In an incomplete statement you will find omissions, whereas the accuracy of a testimony can be undermined by commission errors or distortions. Commission errors are the opposite of omission errors: a witness states that he has seen things that were not there in reality. Distortion means that the facts are being perverted. The witness states, for example, that he saw a green car, when in reality the car was red. The studies I described in my thesis show amongst others that most witnesses give a relatively accurate description. And if there is something lacking in the testimony, it concerns the completeness. These accounts may very well be inconsistent, but that does not make them unreliable." The research she will perform with the veni subsidy focuses on one specific type of memory errors: commission errors. Candel: "The visual memory has the strange characteristic that it extends the boundaries of an observed scene. If a person is shown a picture and later on, he draws this picture from memory, then the image in the picture is always drawn smaller. This means that people remember more background than can be seen in reality. This phenomenon is called Boundary extension. Sometimes people even fill this background with objects that were not at all in the original picture. In that case, we speak of a commission error."

So boundary extension almost always occurs in the human memory, but that does not always involve commission errors. Ingrid Candel is going to study which circumstances cause the development of commission errors. Of course, she already has a hypothesis: the perceptual scheme hypothesis. Candel explains: "When someone sees a certain image, that image can evoke an association with similar images in the memory. In other words: perceptual schemes are activated. A photograph that portrays a horizontal face with a major head wound, for example, evokes associations with an injured person in a hospital bed. When people draw this scene later on from memory, the face is not only drawn much smaller, but sometimes the picture is even supplemented with all kinds of typical hospital furniture." Ingrid Candel expects that this phenomenon occurs sooner if the witnessed scene is more complex. The more neutral and trivial the image, the smaller the risk of commission errors.

In the veni research she will study this hypothesis. Not in real life, but in a laboratory situation. "Only then can you measure the accuracy of recollections. If you ask people to describe from memory something they actually experienced, you cannot compare that with the real situation, because who knows what really happened? Even if you were present at the event, you cannot assess the accuracy, because your own memory is not completely reliable either. Furthermore, it is difficult to influence the complexity of a scene in real life. That is why we will carry out this research with, amongst others, the above drawing tasks."

Together with a.o. professor Harald Merckelbach (professor of Experimental Psychology at UM), Ingrid Candel already regularly published about the Maastricht research into reliability of eyewitness accounts. Their research is important for the field of psychology and law. "Thanks to our publications, we are more and more often called in as expert witnesses by lawyers and the Public Prosecutor to judge the reliability of a testimony or statement of the alleged victim. I have been involved several times in vice cases, where I analyse the statement of the alleged victim to find the reasons for possible inconsistencies and contradictions. The fact is that these do not always have to point to an unreliable and therefore unusable account."

Dr. Ingrid Candel works in the department of Experimental Psychology of the Faculty of Psychology. E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it. . The research into the reliability of eyewitness accounts comes under the activities of the UM Workgroup Psychology & Law.

Rate this item
(0 votes)
Read 31807 times
You are here: Home

Maastricht University Webmagazine

Marketing & Communications
Postbus 616, 6200 MD Maastricht
Minderbroedersberg 4-6, 6211 KL  Maastricht
Tel: +31 43 388 5222
This email address is being protected from spambots. You need JavaScript enabled to view it.

Download UM Magazine

Read a digital version of Maastricht University magazine,
or download the PDF.
UM Webmagazine June 2017

Connect with us