Reconstructive Memory Theory
LOFTUS AND PICKRELL (1995)
RONALD COTTON
LOFTUS AND PICKRELL (1995)
click to edit
DESCRIPTIONS
click to edit
EVALUATION
click to edit
ERRORS AFTER MEANING
Studies have been replicated and show a high degree of reliability
There is evidence that what is seen in the laboratory is seen in real life. (I.e Holocaust testimonies recalling events)
Research has been applied to improve the process of data gathering from witnesses. Additionally helps with understanding false memories
Many say studies lack ecological validity
Studies by Loftus under controlled conditions are open to criticism, as they are artificial in nature (i.e being in an actual accident compared to labs)
Ethical concerns about manipulating participant’s memories. (I.e deception in the Lost in the Mall study)
Schema influences what is encoded and retrieved from memory. Some information might be dropped to streamline memory processing
Based on the idea that memories are not saved as complete, coherent wholes. Instead it is saved as points of data about the object or event
Retrieval of memory is influenced by perception, beliefs, experience, cultural factors, and context in which we are recalling
Schema links to reconstructive memory as it provides data points for our memory to use
click to edit
Bartlett argued that we try to make sense of the past by adding our interpretations of events and deductions of what happened
He argued that memory is an imaginative reconstruction of experience
Aim
Participants
Procedure
To determine if false memories of autobiographical events can be created through the power of suggestion
3 males, 21 females
Before the study, a parent or sibling was contacted and asked two question:
Participants received a questionnaire in the mail. It asked the participants to write about 4 memories then mail back the questionnaire
Three events were real, one was “getting lost in the mall”
Instructed that if they do not remember the event, write “I do not remember this”
Participants were interviewed twice over a period of 4 weeks. They were asked to recall as much information as they could about the events
They were then asked to rate their level of confidence about the event from 1 - 10
After the second interview, they were debriefed and asked to guess which memory was fake
"Could you retell three childhood memories of the participant"
"Do you remember a time when the participant was lost in a mall"
click to edit
click to edit
Evaluation
Results
25% of the participants “recalled” the false memory. However, they ranked this memory as less confident than other memories
Reasoning for the false memories is thought to be because of the schema or stereotype of being lost in a mall (data points of being lost in a mall)
Although this is often seen as strong evidence of the power of suggestion, only 25% had them
The study does not tell us why some participants were more susceptible than others
It was possible to verify the memories through the involvement of parents and siblings
Ecological validity was high as people were talking about childhood memories
The research was applied in areas of eyewitness testimony and therapy
It’s difficult to know whether this is “true” false memory or distortion of another
Ethical concerns about deception about making the participant believe in a false memory
The questionnaire could be contaminated as it was filled at home
There could be demand characteristics, such as social desirability
EXPERIMENT 2
EXPERIMENT 1
Participants
Procedure
Aim
Aim
click to edit
Investigate whether the use of leading questions would affect estimation of speed.
45 student participants
Participants were asked about the speed of the car in different ways. (i.e How fast were the cars when they smashed/hit/collided with each other?)
Questions were based on the assumption “hit” and “smashed” have different connotations and schemas
Researchers predicted the word “smash” would result in higher estimation
IV were the words and DV was the estimation of speed
Students were divided into 5 groups of 9 students each.
Participants watched a total of 7 films of traffic incidents taken from the driver's education films with lengths ranging from 5-30 seconds.
They were asked to give an account of the accident and answer a questionnaire with different questions, with a critical question being to estimate the speed
The critical questions had different words depending on the group. The words were “hit”, “collided”, “smashed”, “bumped”, “contacted”
click to edit
Results
Evaluation
Smashed - 40.8 mph
Bumped - 38.1 mph
Hit - 34.0 mph
Contacted - 31.8 mph
Collided - 39.3 mph
Experiment was a lab experiment and participants were students
This means variables were controlled but a low ecological validity
Films were made for teaching purposes and therefore participants did not receive the same emotions they would have for a real accident.
Confounding variables are controlled so only the effect of the independent variable is measured.
Students are a small sample size, and were most likely young and inexperienced drivers, which may have affected their estimation.
Most people would have a problem estimating the speed of a car
Participants
Procedure
Investigate if participants who estimated high in experiment 1 would say the saw broken glass in the second experiment. Researchers hypothesized that this would happen
150 student participants
150 students were divided into 3 groups of 50 students each.
Participants watched one minute of film containing clips of car accidents all about 4 seconds each.
Participants were asked to describe the accidents in their own words and answer a few questions about the film they just watched.
A week later the participants came back to answer 10 questions about the film.
A critical question was “Did you see any broken glass?” with a yes or no answer
There was no broken glass in the video, but researchers assumed broken glass was associated with high speed
click to edit
Group 2 was asked the same but replaced smashed with hit
Group 3 was not asked this question and became the control group
Group 1 was asked “How fast were the cars going when they smashed into each other?”
click to edit
click to edit
Results
Smashed: 10.46 mph
Hit: 8.00 mph
Control
16 Yes
7 Yes
34 No
43 No
6 Yes
44 No
Evaluation
Results can be interpreted in Bartlett’s theory where people change details when trying to remember things.
This is probably what happened to the participants when they were given information through the key words of either “smashed” or “hit”
Participants may have used past knowledge of car accidents to make decisions on broken glass (schema)
Study can lack ecological validity and therefore may be difficult to generalize findings
Question
Goal
Resolution
Why was the victim so certain that Ronald Cotton was her rapist?
To have a better understanding of why this happened
Important Details
Jennifer Thompson, 22 year old college student, was raped by Bobby Poole on July 28, 1984
Jennifer made a deliberate effort to study his physical characteristics
She survived, and reported the incident. During the investigation, a composite sketch was made
Ronald Cotton was given as a tip. He also worked nearby and had a criminal record
Given photos of suspects, Jennifer confirmed Ronald Cotton was the rapist with certainty
Even after a physical lineup, Jennifer was still sure it was Ronald Cotton
Apparently, Ronald Cotton looks extremely similar to Bobby Poole, who admitted to his crimes in prison
DNA evidence freed Ronald Cotton
During the photos and lineup, they all had different clothing
Follow-up Notes
Research shows memory is highly malleable, and witnesses who begin with uncertainty become more certain over time
Officer was present during the identification, and could've subconsciously provided information
Jennifer's confidence was influenced by the feedback from her choice in the lineup.
Repeating only one individual in multiple procedures or lineups increases witness confidence
Witnesses are more likely to pick someone in clothes similar to those worn by the culprit than physical characteristics, therefore, they should all be wearing the same clothing
All members of the line-up should match the suspect description. In addition, witnesses should be told the suspect may or may not be in the line-up
When gathering evidence from a witness, researchers use a narrative interview style called a Cognitive Interview
Culter & Penrod advocate sequential line-ups. Accuracy of identification increases when suspects are seen one-by-one, and identification is made (yes/no) after each person is presented. Witnesses should not be given feedback that confirms their identification
A narrative interview is an interview that asks simple questions (“Could you tell me what you remember about the night?”). The interviewee does most of the talking, with very few questions except for clarification. This way, interviewer does not alter schema and distort memory by asking leading questions
The cognitive interview begins with context reinstatement. We have better recall when we are in the same place, emotional state, and/or same context in which memory was encoded
This is based on Tulving & Thomson Encoding Specificity Hypothesis (1973). Before asking them to recount, the police would have the interviewee think about where they were and how they felt at the time.
Cognitive interview often uses the following strategies:
Change perspective:
Involves asking the person to “think outside of their schema” (i.e what do you think that the bank teller saw?)
Change the order:
Breaks down the role of schema in “filling in” information. Researchers found that more information is obtained if the witness is asked to recall events forward and backward than simply retelling the story