This investigation explores the subject area of reconstructive memory that suggests the lack of reliability on memory at a cognitive level of analysis. The influential ability of external factors hinder the accuracy of the representation of our experiences. Eye witness testimony is a legal account given by a person who has witnessed an event by sight. The theory of reconstructive memory validates that memory is often flawed and false and it is not an exact replication of the witnessed event. It may be modified abiding by the witnesses’ personal background and extracted information of an event.
According to Bartlett (1932), reconstructive memory was a theory that was pivotal to understand the posited influence of personal interpretation that the ability to recall was subjected to. These interpretations root from our inherited or learnt beliefs, values and cultural norms that shape our perception of our surroundings, therefore leading us to questioning the reliability of eye witness testimony. Therefore, the missing pieces of information during the process of retrieval of memory are filled up significantly based upon their schema. The schema theory developed by Bartlett rapidly grew crucial in the field of criminology ad eye witness testimony. He had tested this investigation through his notable ‘War of the Ghosts’ experiment which was further supported by Allport and Postman (1947). They aimed at validating the theory and added credibility by testing the reliability of memory and the effect of an individuals schema on the their interpretation of an event.
Later, these finding were criticised and contradicted by Yuille and Cutshall (1986). Their study aimed at investigating the eye witness testimony on a real life event of a robbery. While both sides to the validation of this theory are justified, Loftus and Palmer have accurately pointed out loopholes n the reliability of eye testimonies.
The original study replicated to aid this investigation, Loftus and Palmer (1974) aimed to investigate how the reconstructive memory of an eye witness for a particular event may be altered and affected by the language used in the information provided to them post – event. The hypothesis stated that the language used in an eyewitness testimony can alter memory. The null hypothesis on the other hand was that the language used in an eyewitness testimony cannot alter memory. They deduced increase in the intensity of the verb was directly proportional to the increase in the resultant of the participants estimated speed. Loftus and Palmer suggested through their conclusions that the estimation of speed from the participants being lower or higher due to the intensity of the verb was subject to the schemas of the vets that they uphold.
Hence in spite the criticism, there is an influence of an individual’s schemas on their ability to accurately retrieve memory. This investigation will attempt to replicate and recreate the conditions of the Loftus and Palmer (1947) experiment as accurately as possibly to verify the results of their study with the aim of ‘investigating the effect of language on the retrieval of memory’.
Null Hypothesis: The verb used in a question about the speed of a car in an accident will have no effect the estimate of speed given by the eyewitnesses and any difference will be due to chance.
Research Hypothesis: The verb used in a question about the speed of a car in an accident will affect the estimate of speed given by the eyewitnesses.
The independent measures design was used in this experiment as it tests two different conditions while eliminating testing errors and order effects, allowing maximum internal validity. This design was chosen over the repeated measures design as it would hinder the ability to compare the effectiveness of the two different verbs posed through the critical question. If the participants were the same, they will be more knowledgable on what is to be remembered upon the completion of the first trial, leaving the results of the second trial manipulated. The standardised instructions and video clip (car accident) shown to both groups was kept constant to avoid the influence of confounding variables such as the inability of participants to drive or their lack of knowledge on the KPH scale of measurement and to eliminate undesired hinderances to the accuracy of the deduced results for the investigation.
The independent variable of the investigation was the intensity of the definably different verbs posed to the participants through the critical question being ‘hit’ or ‘smashed’ that manipulated the dependant variable which was the participant’s speed estimate of the accident aiding in procuring interval data in Km/h measurement. It was operationalised by creating an observable environment where the groups could be categorised by the definable verb posed in their critical question and the measurable variable through the group they were placed in – ‘hit’ or ‘smashed’. Having two experimental groups helped reveal the influence on participant’s responses by the manipulation of language.
In order to meet ethical guidelines, the participants were given informed consent that informed the participants of the nature of the investigation, eliminated issues of anonymity and deception while allowing them to withdraw their participation. Furthermore, they were debriefed at the start and post-investigation on the purpose and procured results while giving them the ability to withdraw their data.
The sample population of 20 participants aged between 16 and 19 years was derived from the target population of high school students at GEMS Wellington Academy – Silicon Oasis, a multi lingual and cultural international school that creates an open mind mindset on the diverse sample population. The sample was selected through opportunity sampling due to time constrains. The sample population was then categorised between the two experimental groups with 10 participants each through a draw of lots that resulted in the ‘hit’ group with 4 females and 5 males while the ‘smashed’ group had 6 females and 4 males.
Materials used for the investigation:
Electronic Device to watch the video – Laptop
Speakers – to aid the auditory senses of the participants
Questionnaire 1 – 10 copies for the participants of the ‘hit’ experimental group
Questionnaire 2 – 10 copies for the participants of the ‘smashed’ experimental group
Debriefing and briefing material
As the investigation used opportunity sampling, the potential participants were given the opportunity to participate in the psychological experiment.
Upon receiving the participant’s approval, they were required to sign the letter of consent. (Appendix 1: Sample consent form)
Participants were categorised into the ‘hit’ and ‘smashed’ experimental group by a draw of lots.
The standardised instructions were read out to the participants of each group. (Appendix 4: Standardised instructions)
The participants were shown the video clip of the car accident in their respective groups.
Each participant received a questionnaire depending on the experimental group they were placed in that stated “How fast were the cars going when they hit/smashed each other?” (Appendix 2: Questionnaire 1 and Appendix 3: Questionnaire 2).
The participants were acknowledged for their commitment and time.
All participant were debriefed. (Appendix 5: Briefing and debriefing statements)
After receiving all the data, they were recorded on a table.
Table 1: Standard deviation and Mean for both experimental groups derived from the raw data
The mean estimate speed was 84Km/h for the ‘Smashed’ experimental group while it was 67Km/h for the ‘Hit’ experimental group. The standard deviation is the measurement of the dispersion of the data and how far they fall from the mean. The standard deviation was found to be higher for the ‘Hit’ experimental group with the numerical value of 11.1 than the ‘Smashed’ experimental group with he numerical value of 9.7. The investigation recognises the possibility of the results being influenced by the fact that the participants were aware of their participation involving a psychology experiment and also had a personal connection with the experimenter which may reflect the hawthorne effect and lead to demand characteristics.
The deduced results has lead to the investigation’s acceptance of the Research Hypothesis: The verb used in a question about the speed of a car in an accident will affect the estimate of speed given by the eyewitnesses.
Figure 1: The effect of the verb used in the critical question on the estimation of speed on car accident
The results procured through this investigation abide by the replicated study, Loftus and Palmer (1947), suggesting that an individual’s schema can influence the ability to retrieve memory and the response given by the participant. The participants behaved as expected aligning to our research hypothesis when posed with a misleading question. The mean peed estimate was significantly higher for the verb with the higher intensity being ‘Smashed’ as opposed to ‘Hit’ which further validates that individuals process information given to them in accordance to their pre existing representation and knowledge. Thus, this investigation leads to concluding that the intensity of the verb communicated to the participant through a lead-in question interfered with the ability to accurately retrieve memory with respect to the speed of the cars in the car accident causing the differentiated results in the two experimental groups.
As Loftus and Palmer obtained similar findings, it validates the argument that the effect of language on memory is significant in eye witness testimonies as it may lead to misleading and inaccurate representations of witnessed events. This has been clearly outlines and demonstrated through both the original study and my investigation.
There are however limitations to the carried investigation. Conversations that may haven’t take place between the participants even upon urging them to remain quite during the briefing may have lead to biased results due to conformity. There is also a lack of ecological validity, similar to the original study for which Loftus and Palmer were criticised. The artificial environment hinders the ability for the participants to witness a more dramatic event and their involvement with a more realistic situation aiding in more concentration on the accident. This aspect makes it difficult to generals the results and findings to real life which validates the criticism posed by Yuille and Cutshall (1986).
A more representative sample would be advisable for future research on this field to eradicate the existing limitations as this investigation was limited to only IB students and the inevitable element of communication between participants may be prevented by placing them in different rooms. Ecological validity is a significant aspect that must be taken into consideration and made more realistic to deduce more accurate findings and to avoid criticism.
McLeod, S. (1970, January 01). Saul McLeod. Retrieved December 18, 2017, from https://www.simplypsychology.org/eyewitness-testimony.html
Cooper, N. (2011, July 2). Loftus and Palmer Replication Crash Footage. Retrieved Dec 16, 2017, from https://www.youtube.com/watch?v=Rg5bBJQOL74
Loftus, E., & Palmer, J. (1974). Reconstruction of Automobile Destruction : An Example of the Interaction Between Language and Memory’. Retrieved Dec 17, 2017, from https://webfiles.uci.edu/eloftus/LoftusPalmer74.pdf
Crane, J., & Hannibal, J. (2009). IB Diploma Programme: Psychology Course Companion. Oxford: Oxford University Press
Law, A., Halkiopoulus, C., & Bryan-Zaykov, C. (2010). Psychology: developed specifically for the IB. Harlow: Pearson.
Law A, Halkiopoulos C, Bryan–Zaykov C. (2010). Psychology. Oxford: Pearson
Loftus EF, Palmer JC. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. J Verb Learn Verb Be 13(5):58589
Appendix i: Informed Consent Letter
You will be participating in a cognitive psychological experiment, a replication and modification of the established Loftus and Palmer (1947) study. During this experiment:
All procured data will be kept confidential
No psychological or physiological harm would be caused
Your identity will remain anonymous
You have the right to withdraw yourself and your consent from this experiment at any point
You have the right to withdraw the your data upon the completion of the briefing and knowledge of the purpose of the study
Participant may request for the results of the experiment from the experimenter upon the completion and data analysis of the investigation.
I, ____________________, acknowledge and understand the statement laid out by the experimenter and hereby give my consent to participate in the experiment. I also understand that I may withdraw both myself and my data at any given time during the course of the experiment.
Appendix ii: Questionnaire 1
Please respond to the following question based on your understanding of the content shown in the video clip of the car accident.
How fast were the cars going when they hit each other?
Appendix iii: Questionnaire 2
Please respond to the following question based on your understanding of the content shown in the video clip of the car accident.
How fast were the cars going when they smashed each other?
Appendix iv: Standardised Instruction (For both groups)
You will shortly be tested on the reconstructive nature of your memory on a video clip of a car crash.
Please refrain from communicating with other participants during the experiment and maintain decorum at all times.
You will not require anything except a pen/pencil throughout the experiment.
You will soon be provided will consent forms that you will be required to sign to confirm your participation before we begin.
Once the video clip has been played, you will receive a questionnaire to complete
Upon the lapse of the given time, your questionnaires will be collected back and the experiment will end with a debriefing session.
Appendix v: Briefing and Debriefing Statements
“Dear participants, I highly appreciate your willingness to participate in my psychology experiment for my Higher Lever Internal Assessment. This investigation will test the nature of you reconstructive memory. You will be required to view a 10 second video clip of a car accident followed by a questionnaire of the content of the video. I will maintain anonymity and confidentiality of your responses. I would like to remind you that you have the right to withdraw yourself and your data at any given time. Thank you for your support and cooperation. you will now be escorted to the testing area.”
“Dear participants, I would like to thank you for your willingness to participate in the psychology experiment. This experiment aimed at investigating the affect of phrasing the critical question on the judgement of your speed estimation. This study was based on Loftus and Palmer’s research on reconstructive memory (1947). You were all placed in two different experimental groups, both having seen the same video clip of the car accident were posed with different questionnaires. Group 1 was asked “How fast were the cars going when they hit each other?” while Group 2 was asked “How fast were the cars going when they hit each other?”. I expected to see a change in speed estimations by changing the intensity of the critical word in the leading question. My hypothesis was that the ‘smashed’ experimental group would have higher speech estimations being a more intense verb in the critical question in comparison to ‘hit’.
Deducing from our results from both groups, the average estimates of speed for the ‘smashed’ experimental group was higher than the average estimates of speed for the ‘hit’ experimental group which validates my proposed hypothesis abiding by the original study conducted by Loftus and Palmer on reconstructive memory.
For more details on the results or any other clarification, feel free to approach me. While appreciating and respecting your privacy, your data and results will remain confidential and anonymous.
Thank you taking part in my experiment. Your support and cooperation is highly appreciated.”