LINKAGES Memory and Perception in the Courtroom

225
Constructing Memories
FIGURE 6 .8
The Effect of Schemas on Recall
Figure shown
to participants
In a classic experiment, people were
shown figures like these, along with labels
designed to activate certain schemas
(Carmichael, Hogan, & Walter, 1932). For
example, when showing the top figure, the
experimenter said either “This resembles
eyeglasses” or “This resembles a dumbbell.” When the participants were later
asked to draw these figures from memory,
their drawings tended to resemble the
items mentioned by the experimenter. In
other words, their memory had been altered by the schema-activating labels.
LINKAGES
How accurate is eyewitness
testimony? (a link to Sensation
and Perception)
Group 2
Group 1
T
Label
given
Figure drawn
by participants
Label
given
Eyeglasses
Dumbbell
Hourglass
Table
Seven
Four
Gun
Broom
Figure drawn
by participants
LINKAGES
here are few situations in which accurate retrieval of memories is more
Memory and Perception in
important—and constructive memory
is more dangerous—than when an eyewitness
the Courtroom
testifies in court about a crime. Let’s consider
the accuracy of eyewitness memory and how
it can be distorted. To a jury, the most compelling evidence a lawyer can provide is that
of an eyewitness, but eyewitnesses often make mistakes (Loftus & Ketcham, 1991; Wells,
Olson, & Charman, 2002). In 1984, for example, North Carolina college student Jennifer
Thompson confidently identified Ronald Cotton as the man who had raped her at
knifepoint. Mainly on the basis of Thompson’s testimony, Cotton was convicted of
rape and sentenced to life in prison. He was released eleven years later, when DNA evidence revealed that he was innocent (and identified another man as the rapist). The
eyewitness/victim’s certainty had convinced a jury, but her memory had been faulty
(O’Neill, 2000).
Like the rest of us, eyewitnesses can remember only what they perceive, and they
can perceive only what they attend to (Backman & Nilsson, 1991). The witnesses’
task is to report as accurately as possible what they saw or heard. But no matter how
hard they try to be accurate, there are limits to how valid their reports can be
(Kassin, Rigby, & Castillo, 1991). For example, hearing new information about a
crime, including in the form of a lawyer’s question, can alter a witness’s memory
(Belli & Loftus, 1996). Experiments show that when witnesses are asked “How fast
were the cars going when they smashed into each other?” they are likely to recall a
higher speed than when asked “How fast were the cars going when they hit each
other?” (Loftus & Palmer, 1974; see Figure 6.9). There is also evidence that an object
mentioned during questioning about an incident is often mistakenly remembered as
having been there during the incident (Roediger et al., 2001). So if a lawyer says that
a screwdriver was lying on the ground (when it was not), witnesses may recall with
great certainty having seen it (Ryan & Geiselman, 1991). This misinformation effect
can occur in several ways (Loftus & Hoffman, 1989). In some cases, hearing new
information can make it harder to retrieve the original memory (Tversky & Tuchin,
1989). In others, the new information may be integrated into the old memory, making it impossible to distinguish from what was originally seen (Loftus, 1992). In still
others, an eyewitness report might be influenced by the person’s assumption that if
a lawyer or police officer says an object was there, or that something happened, it
must be true.
226
FIGURE 6 .9
The Impact of Questions on
Eyewitness Memory
After seeing a filmed traffic accident, people were asked, “About how fast were the
cars going when they (smashed into, hit,
or contacted) each other?” As shown here,
the witnesses’ responses were influenced
by the verb used in the question. “Smashed”
was associated with the highest average
speed estimates. A week later, people who
heard the “smashed” question remembered the accident as being more violent
than did people in the other two groups
(Loftus & Palmer, 1974).
Chapter 6
Question
About how fast
were the cars going
when they
each other?
Original information
Verb
Estimated mph
smashed into
40.8
hit
34.0
contacted
30.8
External information
Memory
The “memory”
About how fast were
the cars going when
they SMASHED
INTO each other?
Jurors’ belief in a witness’s testimony often depends as much (or even more) on how
the witness presents evidence as on the content or relevance of that evidence (Leippe,
Manion, & Romanczyk, 1992). For example, many jurors are particularly impressed by
witnesses who give lots of details about what they saw. Extremely detailed testimony
from prosecution witnesses is especially likely to lead to guilty verdicts, even when the
details reported are irrelevant (Bell & Loftus, 1989). Apparently, when a witness reports
details, such as the exact time of the crime or the color of the criminal’s shoes, jurors
assume that the witness had paid especially close attention or has a particularly good
memory. This assumption seems reasonable, but there are limits on how much people
can pay attention to, particularly when they are emotionally aroused and the crime happens quickly. Witnesses whose attention was drawn to details such as shoe color might
not have had time to accurately perceive the criminal’s facial features (Backman &
Nilsson, 1991). So the fact that an eyewitness reports many details doesn’t guarantee
that all of them were remembered correctly.
Jurors also tend to believe witnesses who are confident about their testimony (Leippe
et al., 1992). Unfortunately, research shows that witnesses’ confidence is frequently
much higher than the accuracy of their reports (Shaw, 1996; Wells et al., 2002). In some
cases, repeated exposure to misinformation and the repeated recall of that misinformation can lead witnesses to feel certain about their testimony even when—as in the
Jennifer Thompson case—it may not be correct (Lamb, 1998; Mitchell & Zaragoza,
1996; Roediger, Jacoby, & McDermott, 1996).
The weaknesses inherent in eyewitness memory can be amplified by the use of police
lineups and certain other criminal identification procedures (Haw & Fisher, 2004; Wells
& Olson, 2003). In one study, for example, participants watched a videotaped crime
and then tried to identify the criminal from a set of photographs (Wells & Bradfield,
1999). None of the photos showed the person who had committed the crime, but some
participants nevertheless identified one of them as the criminal they saw on tape. When
these mistaken participants were led to believe that they had correctly identified the
criminal, they became even more confident in the accuracy of their false identification
(Semmler, Brewer, & Wells, 2004; Wells, Olson, & Charman, 2003). These incorrect, but
confident, witnesses became more likely than other participants to claim that it had
been easy for them to identify the criminal from the photos because they had had a
good view of him and had paid careful attention to him.
Since 1973, at least 123 people, including Ronald Cotton, have been released from
U.S. prisons after DNA tests or other evidence revealed that they had been falsely
convicted—mostly on the basis of faulty eyewitness testimony (Death Penalty Information Center, 2006; Scheck, Neufeld, & Dwyer, 2000; Wells et al., 2000). DNA evidence