232 • CHAPTER 8 Everyday Memory and Memory Errors
a person from a lineup even when the perpetrator is not present. When a witness
assumes the perpetrator is in the lineup, this increases the chances that an innocent
person who looks similar to the perpetrator will be selected. In one experiment, tell-
ing participants that the perpetrator may not be present in a lineup caused a 42 per-
cent decrease in false identifications of innocent people (Malpass & Devine, 1981).- When constructing a lineup, use “fillers” who are similar to the suspect. Police
investigators are reluctant to increase the similarity of people in lineups because
they are afraid this will decrease the chances of identifying the suspect. However,
when R. C. L. Lindsay and Gary Wells (1980) had participants view a tape of a
crime scene and then tested them using high-similarity and low-similarity lineups,
they obtained the results shown in ● Figure 8.23. Figure 8.23a shows that when
the perpetrator was in the lineup, increasing similarity did decrease identification
of the perpetrator, from 0.71 to 0.58. However, Figure 8.23b shows that when the
perpetrator was not in the lineup, increasing similarity caused a large decrease in
incorrect identification of an innocent person, from 0.70 to 0.31. Thus, increasing
similarity does result in missed identification of some guilty suspects, but substan-
tially reduces the erroneous identification of innocent people, when the perpetrator
is not in the lineup. - When presenting a lineup, use sequential rather than simultaneous presentation.
The usual depiction of lineups in movies—and the one most often used in police
work—is 5 or 6 people standing in a line facing the witness, who is hidden behind
a one-way window. The problem with this way of presenting a lineup is that it
increases the chances that the witness will make a relative judgment—comparing
people in the lineup to each other, so the question is “Who is most like the person
I saw?” However, when each person in the lineup is presented sequentially—one at
a time—then the witness compares each person not to the other people, but to the
memory of what the witness saw. Lindsay and Wells (1985) found that for lineups
in which the perpetrator was not present, an innocent person was falsely identified
43 percent of the time in the simultaneous lineup, but only 17 percent of the time in
the sequential lineup. The beauty of the sequential lineup is that it does not decrease
the chances of identifying the suspect when he or she is included in the lineup. - Use a “blind” lineup administrator and get an immediate confidence rating. When pre-
senting a lineup, the person administering the lineup should not know who the suspect
is. In addition, having witnesses immediately rate their confidence in their choice elimi-
nates the possibility that the postevent feedback effect could increase their confidence.
5. Improve interviewing techniques. We have already
seen that making suggestions to the witness (“Good,
you identified the suspect”) can cause errors. Cogni-
tive psychologists have developed an interview pro-
cedure called the cognitive interview, which is based
on what is known about memory retrieval. This inter-
view procedure involves letting the witness talk with
a minimum of interruption and also uses techniques
that help witnesses recreate the situation present at
the crime scene by having them place themselves back
in the scene and recreate things like emotions they
were feeling, where they were looking, and how the
scene may have appeared when viewed from different
perspectives. Comparisons of the results of cognitive
interviews to routine police questioning have shown
that the cognitive interview results in 25–60 percent
more information than the usual police interview
(Fisher et al., 1989; Geiselman et al., 1985, 1986).
Recommendations like those described above led
to the publication in 1999 of Eyewitness Evidence:
A Guide for Law Enforcement by the U.S. Justice●FIGURE 8.23 Results of Lindsay and Wells’ (1980) experiment,
showing that (a) when the perpetrator was in the lineup, increasing
similarity decreased identifi cation of the perpetrator, but (b) when the
perpetrator was not in the lineup, increasing similarity caused an even
greater decrease in incorrect identifi cation of innocent people.Low
SimilarityHigh01.00.5Proportion witnesseschoosing guilty suspect(a) Perpetrator in lineupLow
SimilarityHigh01.00.5Proportion witnesses
choosing innocent person(b) Perpetrator not in lineupCopyright 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part.