Reliability of Eyewitness Testimony: A Field Experimental Approach for a Real Crime

Abstract

To develop a method to assess the reliability of eyewitness testimony in a real criminal case, a field experimental or simulation method was introduced. In that case, an eyewitness witnessed a man at night who was in an escaping vehicle and identified him as a criminal from a photo-spread about two weeks later. The description of the eyewitness was detailed; however, there might be estimator variables that lower his performance of perception and memory. To clarify whether such testimony is trustworthy, a field experimental method was adopted. Results of the experiment showed that almost all subjects of the experiment could not only not recall the details of the events but also could not identify the target person. From these findings it is concluded that the testimony was not reliable enough to convict the suspect. In this paper the usefulness of a field experimental approach is stressed and it is suggested that guidelines such as PACE or recently issued US guidelines for treating eyewitness testimony should be used to prevent miscarriage of justice.