2.7 Misleading Questions - Factors affecting accuracy of EWT
Leading Questions
Loftus and Palmer (1974)
Procedure: Had 45 participants watching clips of car accidents
then asking questions about the accident. In the critical question,
participants had to answer how fast the car was going. 5 groups
of participants and each group was asked the question using a
different verb : hit, contacted, bumped, collided, smashed
Findings: Mean speed for contacted was 31.8 mph
whereas the verb smashed had 40.5 mph mean.
Response-bias explanation
Suggests that the wording of the question
has no real effect on the memories but just
influences how they decide to answer.
Conducted a second experiment in which
participants with the word smashed were
more likely to report seeing broken glass
when there was none.
Post-Event discussion
Gabbert et al (2003)
Procedure: Studied participants in pairs,
watching a video of a crime but from
two different perspectives. This
means each participant could see
elements of the event the other
couldnt. They then discussed what
they had seen.
Findings: 71% of participants mistakenly
recalled aspects of the event they did not
see in the video but discussed.
Memory contamination
When discussing information, their
memories may become altered or distorted.
They combine misinformation from other
witnesses.
Memory conformity
Participants may go along
with information in order
to win the approval of
other people.
Evaluation
Strengths:
Real world application. Important uses in the
criminal justice system, as consequences of
misleading questions can be very serious
Weaknesses:
Evidence against substitution. EWT is more accurate for some
aspects of an event than others. Presumably the participants'
attention was focused on central features. Suggests that the
original memories survived and were not distorted.
Evidence challenging memory conformity.
Skagerberg and Wright (2008) showed their
participants two versions of a film, their views
rather than being changed blended together.