ChaPteR 8 Memory 273then heard a description of the face supposedly
written by another witness—a description that
wrongly said the man had light, curly hair (see
Figure 8.1). When the students reconstructed
the face using a kit of facial features, a third
of their reconstructions contained the mislead-
ing detail, whereas only 5 percent contained it
when curly hair was not mentioned (Loftus &
Greene, 1980).
What if several eyewitnesses are involved?
Does that affect a person’s susceptibility to
misleading information? In a study of this ques-
tion, people first watched a video of a simulated
crime and later read three eyewitness reports
about the crime, each report containing the
same misleading claim—for example, about the
location of objects, the thief’s actions, or the
name on the side of a van. One group was told
that a single person wrote all three reports,
whereas another group was told that each report
was written by a different person. It made no
difference; a single witness’s report proved to
be as influential as the reports of three different
witnesses. Such is the power of a single witness’s
voice (Foster et al., 2012).
Leading questions, suggestive comments, and
misleading information affect people’s memories
not only for events they have witnessed but also
for their own experiences. Researchers have suc-
cessfully used these techniques to induce people
to believe they are recalling complicated events
from early in life that never happened, such as
getting lost in a shopping mall, being hospitalized
for a high fever, being harassed by a bully, get-
ting in trouble for playing a prank on a first gradeOn TV crime shows, witnesses often identify a criminal
from a lineup or a group of photos. But these methods
can mislead witnesses, who may wrongly identify a
person because he or she resembles the actual culprit
more closely than the other people in the lineup or
photos do. Because of psychological research, many
law enforcement agencies are now using better meth-
ods, such as having witnesses look at photos of sus-
pects one at a time without being able to go back to a
previous one.essentials of an experience and then use our
knowledge of the world to figure out the specif-
ics when we need them. But precisely because
memory is so often reconstructive, it is also vul-
nerable to suggestion—to ideas implanted in our
minds after the event, which then become associ-
ated with it. This fact raises thorny problems in
legal cases that involve eyewitness testimony or
people’s memories of what happened, when, and
to whom.
The Eyewitness on Trial LO 8.3
Without the accounts of eyewitnesses, many guilty
people would go free. But, as Jennifer Thompson
learned to her sorrow, eyewitness testimony is not
always reliable. Lineups and photo arrays don’t
necessarily help because witnesses may simply
identify the person who looks most like the per-
petrator of the crime (Wells & Olson, 2003). As
a result, some convictions based on eyewitness
testimony, like that of Ronald Cotton, turn out to
be tragic mistakes.
Eyewitnesses are especially likely to make
mistaken identifications when the suspect’s eth-
nicity differs from their own. Because of unfa-
miliarity with other ethnic groups, the eyewitness
may focus solely on the ethnicity of the per-
son they see committing a crime (“He’s black”;
“She’s white”; “He’s an Arab”) and ignore the
distinctive features that would later make identi-
fication more accurate (Levin, 2000; Meissner &
Brigham, 2001).
In a program of research spanning four de-
cades, Elizabeth Loftus and her colleagues have
shown that memories are also influenced by the
way in which questions are put to the eyewit-
ness and by suggestive comments made during
an interrogation or interview. In one study, the
researchers showed how even subtle changes in
the wording of questions can lead a witness to
give different answers. Participants first viewed
short films depicting car collisions. Afterward, the
researchers asked some of them, “About how fast
were the cars going when they hit each other?”
Other viewers were asked the same question, but
with the verb changed to smashed, collided, bumped,
or contacted. Estimates of how fast the cars were
going varied, depending on which word was used.
Smashed produced the highest average speed esti-
mates (40.8 mph), followed by collided (39.3 mph),
bumped (38.1 mph), hit (34.0 mph), and contacted
(31.8 mph) (Loftus & Palmer, 1974).
Misleading information from other sources
also can alter what witnesses report. Consider
what happened when students were shown the
face of a young man who had straight hair,