Many eyewitnesses are called to testify because they have witnessed a crime, accident or incident. The anxiety if this may cause an affect on the reliability of their EWT. There are two main issues in this question: firstly the prediction of the Yerkes Dodson Law that as arousal increases, then so does performance weather spotting, stage performance, or memory encoding, up to an individual optimum level. However, after this optimum level it is suggested that the performance of the individual will decline, this could be triggered by terror. This theory was supported by Deffenbacher et al’s research. The second issue is the “weapons focus effect”; In Easterbrook’s research he predicted that attention will narrow to the source of the threat e.g the knife the man is holding therefore peripheral detail is lost e.g the details of the perpertrator’s face. This was supported by Loftus’s laboratory studies in an artificial environment using film and slides, using independent groups design, where a control conditions showed the participants a similar scene without weapons.
In the first group they had a scene which they believed to be a real life violent crime involving a weapon, results showed that the participants remembered the knife in great detail but no other features, such as the perpetrator’s face clothing etc. However the participants in the control conditions witnessed the same person but only in a peaceful situation and the participants were able to recognise the man when given 50 photos. When we examine real life eye witness testimony recall however, there is excellent recall of detail, and the weapons-focus effect was not supported. Yullie and Cutshall (1986) conducted a natural experiment on 13 out of 21 bystanders who had witnessed a violent shooting 4-5 months previously. There recall was detailed and accurate and they were resistant to leading questions months after the event.
Therefore, this suggest that bystanders had reached the optimum level ( Yerkes Dodson Law) when they they witnessed the death of the robber who had previously wounded the owner in Vancouver gun shop before the owner killed him. Weapon focus did not influence witness recall. Therefore we need to examine witnesses to the violent crime who faced a weapon rather than a bystander to confirm the effect of weapon focus shown by Loftus. The natural experiment conducted by Christanson and Hubinette (1993) in Sweden re-interviewed 58 witnesses (customers and bank tellers) to bank robberies that occurred around Stockholm in one year. Those witnesses who faced the weapon (bank tellers) they had better recall and accuracy of the event after a period of time rather than the bystanders. Optimum arousal is suggested but weapon focus is challenged.
The ecological validity of this evidence is high although the participants cannot be randomly assigned to the experiment and there is no control condition as there would be in a laboratory experiment. However, Laboratory experiments lack the intense arousal of a real crime so the supposed weapons effect may be the effect of distinct and unexpected events distracting participants to the unusual element in the familiar. A strong point of Loftus’s field experiment and supports the weapon focus effect; is that it has ecological validity and comparison to the control condition provide strong support. A criticism is the inability of researchers fro randomly allocate participants in this study to experimental and control conditions to control individual differences.