Eyewitness Testimonies and the
Reconstructive Nature of Memory:
An experiment on the effects of misleading external information on
the recalling of an event
hbc647
Group Members:
• hbc664
• hbc669
• hbp218
Submission Date: December 15, 2019
Word Count: 2,198
,IB Psychology IA hbc647
CONTENTS
Section Page
I. Introduction 2
II. Exploration 4
1. Procedures
2. Design
3. Sampling Method
4. Participants
5. Controls
6. Materials
III. Analysis 6
1. Results
2. Table
3. Inferential Statistics
4. Graph
IV. Evaluation 8
1. Discussion
2. Strengths
3. Limitations and Modification
V. References 10 - 11
VI. Appendices 12-20
1. Raw Data 12
2. Inferential Statistics 13
3. Informed Consent Form 14
4. Debriefing Form 15
5. Standardized Instructions 16
6. Questionnaire (For “collided” condition) 17
7. Questionnaire (For “smashed” condition) 18
8. Questionnaire (For control) 19
9. Traffic Car Crash Video 20
Tip: you can use a table to create a good layout and then select “no border” to
make it disappear
1
,IB Psychology IA hbc647
I. Introduction
For decades, the accuracy of eyewitness testimonies has been under question. While eyewitness
testimonies are still considered as a major form of evidence by the current juridical system,
studies have shown that eyewitness testimonies can be influenced by many external factors
including post-event misinformation (Eakin, Schreiber, & Sergent-Marshall, 2003; Mahé,
Corson, Verrier, & Payoux, 2015), social influence and personal beliefs (Wilson & French,
2014) as well as schemas (Tuckey & Brewer, 2003). In addition to the practical relevance of the
topic, its theoretical significance to researchers’ understanding of memory processes cannot be
overlooked.
Leading questions are questions that are phrased to encourage a desired answer and are
frequently used when questioning eyewitnesses. Leading questions often contain misleading
information (Bowles & Sharman, 2014). According to Ayers and Reder (1998), long term
memory is very susceptible to inaccuracy and incorrect details may be implemented during
memory reconstruction. The phenomenon of having external information alter memory is known
as the misinformation effect. This can be explained by Bartlett’s (1932) theory of reconstructive
nature of memory. According to Bartlett, memory is affected by prior knowledge, experiences,
and information received. Bartlett’s theory suggests that memory is actively reconstructed as
they are recalled. Thus, memory may not be an accurate representation of reality.
Loftus and Palmer’s (1974) first experiment, the study being replicated, aimed to investigate how
misinformation in leading questions influenced a witness’ memory. The participants were
undergraduate students (n=45), and they were shown segments of traffic accidents. Participants
were then asked a critical question, which asked to approximate the speed of cars, designed to
mimic the leading questions frequently used in questioning. The verb - smashed, collided,
bumped, hit, or contacted - that was used to describe the action of the cars was the independent
variable. The speed estimations were the dependent variable. The results showed 30% higher
speed estimate when the verb “smashed” was used compared to “contacted”.
The results suggested that external misleading information affected the recalling of memory.
Descriptive verbs used in the questions suggested the severity of the crash, which was the
misleading information. This information then affected the participants’ recalling of the
accidents.
The aim of the present study is to investigate whether the use of a descriptive verb indicating a
greater severity of the crash in the critical question would influence the recalling of the event. By
testing this, the study attempts to investigate the theory of reconstructive nature of memory. As
mentioned above, this topic is of significant practical and theoretical importance. By showing
that human memory has faults, convictions based entirely on eyewitness testimonies may be
re-examined to avoid convicting innocent people.
2
, IB Psychology IA hbc647
Independent Variable: Descriptive verbs (smashed or collided) used to describe the car crash in
the critical question.
Dependent Variable: The estimation of speed given by the participants.
Research Hypothesis (H1): There will be a significant difference in speed estimate values
between the three conditions, with “smashed” condition having the highest speed estimate,
“collided” condition having the second highest, and control having the lowest.
Null Hypothesis (H0): There will be no significant difference in speed estimate values between
the three conditions.
3
