IB Docs (2) Team
ERQ marking: Reliability of memory
Below you will find three sample ERQs for the question: To what extent is one cognitive process reliable?
For each of the samples, refer to the rubric to award marks. After each sample, there is a predicted grade as well as feedback on the strengths and limitations of the sample.
ERQ rubricSample 1
Although we all seem to trust our own memories, many psychologists argue that our memories are less reliable than we think. Some memories, such as procedural memories, seem to be reliable. We never forget how to ride a bike or how to swim. However, episodic and semantic memories, both of which are encoded by the hippocampus, are open to memory distortion.
Loftus carried out a study about the reliability of episodic memory and the ability to implant false memories. She interviewed the parents of her participants to get stories from the participants’ childhood and to find out whether they got lost in a mall when they were children. She then sent a questionnaire to the participants asking them to explain the four events – one being the fake “lost in the mall” story. She then interviewed the participants about their memories. When she asked her participants to describe in detail when they got lost in a mall as children, 25% of the participants recalled the experience, even though the event never happened.
This study has been supported by research by Shaw that shows that people can develop false memories about a crime that they commented, simply by being told that their parents had told the researcher the story and including a few key details. Results appear to be reliable. In addition, the study has high ecological validity. There are, however, some ethical concerns about the manipulation of a person’s personal memories for research purposes. But the results of this research has been applied to the justice system to challenge the idea that eyewitness testimonies should be seen as solid evidence.
In a second study, Loftus showed that information that is received after an event can influence how we remember something. Loftus showed a group of students a series of film clips which included a car accident. All participants were given a questionnaire, but they were in three different conditions. All of the questions were the same, except one of the questions was asked to different participants with different verbs with different intensities –how fast was the car going when it smashed into or hit the other car. One group was not asked this question. Participants were then called back a week later and asked whether they remembered broken glass at the scene of the accident. She found that more participants in the “smashed” condition remembered glass than either the control or the “hit” condition.
This study showed that memory may be influenced by information “after the fact”. But this study lacks ecological validity; in addition to the fact that it was done under highly controlled conditions, the study lacked any emotional change in the participants that would have been experienced if they saw a real accident. The question is what happens in a real life situation?
To find out, Yuille and Cutshall looked at the reliability of memory in eyewitnesses of a robbery in Vancouver. They asked the witnesses a series of questions about the crime, using a technique similar to Loftus. In their study they asked the participants to tell their story; then they asked one of two leading questions – either “did you see a broken headlight on the getaway car?” or “Did you see a yellow panel on the car?” The researchers found that the participants remembered a lot of accurate detail from that day. They could verify it by using the archived police reports. Also, the leading questions had no significant effect on the responses of the participants.
This is study is a bit problematic in that it cannot be replicated. It also is not generalizable as it was a single incident and the people in the sample are not representative of a larger population. Also, because of the time that passed between the event and their retelling of their stories, it is a question how many other variables influenced the accuracy of their stories – for example, the number of times they retold the story or had read information about the crime. But unlike Loftus and Palmer’s study, this study includes emotion. The participants safety was threatened during the robbery This may have resulted in a “flashbulb memory” that was more accurate that the memories created in the Loftus & Palmer experiment.
Neisser and Harsch also carried out a study of the reliability of memory by having participants write down what they remembered about the Challenger disaster right after it happened. The students were asked to write down personal details such as where they were when they heard the news or who was with them. Two and a half years later they were again asked to fill in the same questionnaire. Although the participants were confident of the accuracy of their responses, the reliability of the memories was low. Perhaps it was not the emotion in Yuille and Cutshall that made the difference – but the life-threatening situation which would surely have triggered the amygdala. Although the Challenger disaster was a surprise, we cannot say that people all had a strong personal reaction to the disaster.
In conclusion, episodic memories are not always reliable. It appears that if there is an emotional response and there is a personal relevance for the event, people are more likely to have a reliable memory. This may be the result of biological factors such as the activity of the amygdala.
Word count: 887
Focus on the question: The response is focused on the question of the reliability of memory. The essay clearly develops an argument. 2 marks.
Knowledge and understanding: There is some understanding of the role of the amygdala in the creation of memory. But otherwise, there is little discussion of the reconstructive nature of memory and the role this may play in many of these experiments. Vocabulary is used appropriately. 4 marks.
Use of research: Several studies are used effectively. They are described, but not always well explained. Using less research but being explaining more carefully what the study teaches us about the reliability of memory would make for a stronger essay. 4 marks.
Critical thinking: There is an attempt at critical thinking, but ideas are often not developed. For example, the studies by Loftus & Pickerell and Shaw are said to be “ecologically valid,” but this is not explained. The evaluation of the studies later in the essay is fairly well developed and linked to the demands of the question. 4 marks.
Clarity and organization: The language use and structure of the essay are very good. 2 marks
Total marks: 16 marks
Predicted grade: 6
Sample 2
The reliability of memory, as a cognitive process, has long been an issue that psychologists have dealt with. The role of emotion in memory is often brought up in this discussion. Are memories that are related to emotion more or less reliable?
Brown & Kulik came up with the Flashbulb Memory Hypothesis. Their research concluded that the more emotion that a person indulges in a memory, the better and more accurate they remember it. The researchers described flashbulb memories like a vivid picture.
Another theory, proposed by Craik & Lockhart, suggested that the deeper a memory is processed, the better it is remembered. This theory is known as the Levels of Processing Theory. A limitation of this theory, however, is that it lacks construct validity. What is the definition of “deeper?” Does it involve emotion as suggested by the Flashbulb Memory Hypothesis? Why are some pieces of information remembered without deep processing?
The other side of the argument argues against reliable memory. These studies suggest that memory is unreliable and additionally, is reconstructive.
The first study is by Neisser, who looked at the tragedy of the Challenger crash. This event was going to be a huge national spectacle where a school teacher was sent into space. The anticipation of this event was immense and the public felt emotionally connected to the school teacher. When the crash happened right after take-off, the whole country was in sorrow. Neisser took this opportunity to examine the reliability of memory related to emotion. So he asked his sample of student participants on their actions during the tragic event. He recorded them. Two years later, he came back to the students and asked them the same question on their actions during the event. The majority of the students’ answered differed greatly. Most students were 100% sure that their answers were correct.
Firstly, we have to evaluate the reliability of the study. The sampling was done by a professor on his own students. This suggests that we cannot generalize the findings to humans as a whole; it lacks ecological fallacy because the sample is too narrow. On the other hand, the study was prospective, meaning it observed change over time in memory. Nevertheless, we can see that this study by Neisser contradicts Brown & Kulik’s Flashbulb Memory Hypothesis as well as Craik & Lockhart’s Levels of Processing Theory.
The implications of this study are that memory is reconstructive; human brains create memory upon experiencing highly emotional or traumatic events. A limitation of this study, however, is again the sample. Students are not very representative of a general human population. Additionally, the immediateness of the question suggests that some students may still have been in shock and not known what they were saying. Demand characteristics may have been present here as well together with social desirability effect. When the researcher asks the students if they were watching the event, the students might think that this is what the researcher is looking for - an affirmative answer. Additionally, it would appear that it’s society’s expectation that students watched the tragedy. It could have been that this was the reason for the majority of positive responses.
To conclude, despite a quite applicable and reasonable set of memory theories, presented by Brown & Kulik and Craik & Lockhart, the actual evidence presented by Neisser is hard to dispute. There are many limitations to the study by Neisser, which make it far from indisputable, but at least when discussing a student sample, memory is not absolutely reliable.
Word count: 586
Focus on the question: The essay lacks clear focus on the question of reliability, even though the question itself is identified in the introduction. 1 mark.
Knowledge and understanding: There are several terms used without any, or with very little explanation – e.g. reconstructive memory, flashbulb memory, levels of processing. The term “reliable” is not used correctly. No clear understanding of why the memories of the students in Neisser’s study were not reliable. 2 marks.
Use of research: Only one study is outlined. The study lacks a clear description for someone that does not know the study. Results are poorly stated. The theories that are mentioned are not explained in any detail. Levels of processing is not highly relevant to the central argument of the essay. 2 marks.
Critical thinking: The focus of the evaluation of the Neisser study is on the sample, which is not highly relevant to the question. There is a mention of the problem of measuring “depth of processing”, but this is not linked to the idea of emotion influencing one’s encoding processes. 2 marks.
Clarity and organization: There is an attempt to address the command term “to what extent,” but the essay lacks development of ideas. 1 mark.
Total marks: 8 marks
Predicted grade: 4
Sample 3
Psychologists have helped to change the approach of law enforcement by showing that eyewitness may not be as reliable as one would think. It appears that our memories may be affected by the misinformation effect, emotion and even our own retelling of the story.
One of the earliest studies of the reliability of memory was done by Bartlett. Bartlett read a native American story called “The War of Ghosts” to a sample of British participants. There were two groups. The first group was asked to repeat the story to another person is a “rumour chain.” As expected, Bartlett found that the details of the story became distorted over time. The second group was asked to repeatedly recall the story over time. In other words, they were only repeating it to themselves. What was interesting was that even though there was no interference from other people, the participants in this group also distorted the story in the same way as the first group. They made the story more in line with their own culture, added emotional details and left out information that was not important.
Another study that was done on the reliability of memory was the study by Brewer & Treyens on office schema. In their study, participants thought that they were waiting for the experiment to begin. However, it had already started. They were asked to wait in an office for 35 seconds. After the 35 seconds, they were called into the other room and asked to recall as many items from the office as possible. Some of the items in the office were “schema congruent” – which means that the participant would expect that they would be there. There were also “schema incongruent” items, which means that the participants would not expect them to be there (for example, a brick). When they were asked to recall the items, they remembered more schema congruent than incongruent items. This shows that memory is not reliable and can be influenced by things in our environment.
Loftus & Palmer carried out a study of the misinformation effect. In their study they had students watch a series of videos involving a car crash. They then gave them a questionnaire which was the same for all participants except for one question. They asked them how fast the car was going when it smashed, hit, or collided into the other car. A week later they called the participants back and asked them about the video. They asked them if they remember seeing broken glass at the scene of accident. (There wasn’t any). The participants who had the strongest verb (smashed), misremembered the glass more often than those with the verb collided.
In spite of the fact that both of these studies show that memory may not be reliable, Yuille and Cutshall found that the misinformation effect did not seem to play a role in a real life situation. Their study was a natural experiment. There had been a robbery in Vancouver in which there had been gunshots. There were several witnesses and many of them had been afraid during the attack. After the robbery, all of the witnesses had given testimony to the police department. This is important because the researchers had archival evidence of their memories, unlike the study by Brown & Kulik. Four months later, the researchers asked the participants to take part in their study. In their experiment, they asked the participants the same questions as Loftus and Palmer. They asked them “Did you see the broken glass?” Even though they were given a leading question, the participants did not make any mistakes. Their memories matched the original archives of their testimonies. It seems that emotion may have played a role in their ability to recall the details of the robbery.
So, what does this mean for the reliability of memory? First, it is important to remember that all of the examples above are episodic memory. Other types of memory may be more or less reliable. It also shows that our memory is influenced by other information – whether it is given to us by someone else, or whether it is in our own thinking processes. Finally, it shows that when there is emotion, we can trust our memory more than when there is none.
Word count: 711
Focus on the question: The essay is clearly focused on the demands of the question. 2 marks.
Knowledge and understanding: The essay demonstrates some understanding of the issue of reliability, but key ideas are not developed. There is no mention of the reconstructive nature of memory or schema. There is also no explanation of the misinformation effect. The term “natural experiment” is incorrect. The study was a field experiment that used a purposive sample. Much of the knowledge and understanding is implicit in nature. 3 marks.
Use of research: Four studies are described. The Brewer and Treyens study is not highly relevant to the question of “reliability” as there was no testing of the accuracy of memory over time. There are some errors in the results of Yuille and Cutshall’s study. 4 marks.
Critical thinking: There is no evaluation of the research. There is a paragraph at the end of the essay that attempts to discuss the question, but it simply summarizes the main points and lacks development. 1 mark.
Clarity and organization: The essay is well organized and addresses the command term. 2 marks.
Total marks: 12 marks
Predicted grade: 5
Twitter
Facebook
LinkedIn