Memory Decay and Distortion 1 & 2 (wk 3) Flashcards
List three factors that can influence Memory between storage and retrieval
- Passage of time (forgetting)
- Intervening knowledge/experience (learning)
- Misinformation (Gaining incorrect information regarding an event
Eyewitness memory is …….
Reconstructive
Hermann Ebbinghaus
- Investigated memory scientifically and systematically exploring how info is stored in the brain and decays over time
- Studied this by presenting participants with nonsense syllables and then asked them to recall these after different time delays
- Results led to the formulation of the Ebbinghaus forgetting curve which suggests forgetting is a non-linear process
The Forgetting Function
Forgetting functions describe how accuracy or discriminability declines as the temporal distance from the event to be remembered increases.
This was one of Ebbinhaus’s findings in developing the forgetting curve
Describe how Memory becomes fragile in recall
Memory is formed through encoding and consolidation.
When we retrieve or recall information from our memory, it moves from a stable, consolidated format into a neurobiologically fragile state which is susceptible to change. As such, when memory is recalled and reconsolidated, the memory trace itself can change
What did Soren Kierkegaard say about Intervening Knowledge/Experience?
“Life is backwards, but must be lived forwards”
This stresses the idea that our memory guides our future actions, and that memory is continually being updated to integrate new and present knowledge
Snyder and Uranowitz 1978 Study on Intervening Knowledge/Experience and Memory
- Participants were presented with a case history about “Betty K”
- After one day or one week from reading the case history, participants were told that Betty was either straight or gay (or a control group was told nothing)
- Then participants’ memory for the case history was assessed
Findings showed the delay before intervening was unimportant but that participants made more label-consistent errors in recall than label-inconsistent errors.
Aka. Intervening knowledge (Betty’s sexuality) attributed for some error in the memory trace regarding the case history
Hindsight Bias
= When knowledge of current events, emotions, or outcomes, bias judgments, and the memory of processes leading up to those judgements.
(ie. I always knew he’d cheat on her)
- This is a product of intervening knowledge/experience infiltrating our memory trace
(eg. the more clues we get someone is a cheater, the more likely we are to think we always had doubts about them)
Hindsight bias can both influence our judgements and our memory for past events
Davis, Lopez, Koyama et al., 2005 study on Intervening Knowledge/Experience
Participants listened to an audio tape of a couple arguing
- Half the participants were told the couple subsequently broke up
- The other half were told the female was murdered, and her bf was a suspect in the murder
Participants were then tested on their memory for the audio tape
When told she was murdered, participants were more likely to accurately remember the threatening statements made by the male, however, were more likely to inaccurately recall aspects of the story (like he hit her and threatened her classmate etc)
(like being a liberal responder in a SDT with more hits and more false positives)
Misinformation and Memory
Misinformation is any information gained after a memory is laid down that was not part of the original event. (aka. Post event information)
Misinformation can reliably interfere with our memory trace
Elizabeth Loftus; Standard Misinformation Procedure
- Event (expose a participant to something)
- After some duration of time mislead the participant by planting inaccurate information
- After another duration of time interview the participant about the original event
eg. Watch red car in an accident –> ask “how fast was the blue car going” –> What colour was the car?
Misinformation Effect
Is a robust memory effect, in which misinformation becomes part of our memory trace
TWA 800 Flight as an example of memory inaccuracy
In the 1900’s TWA Flight 800 left New York and crashed shortly after near Long Island
There were no survivors in the crash but many eyewitnesses.
One eyewitness reported seeing something ‘hit’ the plane and media speculated it was a missile which caused the crash
Speculation in media and eyewitness reports were everywhere, until people were convinced they saw the missile hit the plane and believed when reports revealed the plane crashed due to a faulty fuel tank that the government were covering up a terrorist attack
What are three sources of misinformation in memory
- Media
- Other witnesses (co-witness discussions)
- Investigators (ie police)
Media as a source of misinformation ; Ei AI Crash, Amsterdam 1995
Shows how media can be both a good investigative tool but also a source of misinformation
- A plane crashed into an apartment at night and there was no video footage
- media speculated what could have happened
- in a survey 6 - 9 months after the crash 50% of people believed they had seen video footage of the crash
This shows how media reports are impactful as people can create memories for information they read
Media and Misinformation; Loftus and Banaji 1989
Showed participants a short video (4 mins) of a robbery and shooting
Then participants were exposed to either a TV report containing misinformation or no misinformation (that was either consistent or inconsistent with the informaiton in the original stimulus video).
Participants were asked to watch the TV report and evaluate the news caster before they were tested for memory on information in the original stimulus video.
Those exposed to a misinformation filled news caster video were more likely to incorporate misinformation . In fact 1/3 of misinformed participants incorporated the misinfo in their reports, whilst control responses were highly accurate
Co-witness discussions; Paterson and Kemp 2006
85% of witnesses engage in co-witness discussions (often before the police arrive)
According to witnesses, the primary purpose of these discussions is to ‘provide/exchange info’
Why are Co-witness discussions bad for the justice process?
“What a court wants is the independent recollection of the individual witness” - Heaton Armstrong 1987
Co-witness discussions means reports are not independent but a collation of multiple witnesses perspectives which may have originally differed
Possible Advantages to Co-witness discussions (Patterson and Kemp, 2005)
- May prompt a witness to recall forgotten details
- May reinforce memory
- May aid recovery from trauma
- May paint a more accurate overall picture
(these advantages are speculated by police generally)
Possible Disadvantages to Co-witness discussions (Pattern and Kemp 2005)
- May contaminate independant recollection (stressed to be of importance by Heaton-Armstrong)
- May weaken the prosecution case
- Witnesses may become unsure of their testimony
- May lead to reports of ‘common things’
- Collusion
Oklahoma Bombing and Co-witness Testimony
In 1995 an apartment building was blown up by a bomb in a rented truck. When tracking back the rental, the owner of the rental company said he remembered loaning the car to two men (however, the two women working in the office said they only remembered one person). Police began looking for two culprits until the rental car owner said he had misremembered and it was only one person who rented the car.
Interestingly, the two women who remembered correctly at the beginning, after hearing their boss admit to only one person having rented the car, felt convinced it was actually two people
Soloman Asch - Conformity Studies
The Solomon Asch paradigm involved presenting a stimulus line, followed by 3 test lines - participants are asked to match the stimulus line to the test line that is of the same length.
In a staged event, participants would do this task in front of a number of confederates, who would intentionally answer wrong (unanimously)
Asch found, participants would alter their response and choose the wrong answer to match the confederates 37% of the time
The conformity response reduced, when there were 2 participants in the group of confederates and when participants were asked to write their answers on paper instead
What are the types of conformity resulting from Solomon Asch’s paradigm
Informational conformity = when a participants gives the wrong answer (the answer that matches confederate responses) as they believe they must be wrong, if the rest of the group is correct
Normative conformity = when a participant knows the answer they’re giving is wrong but they don’t want to disagree with the group
Based from Asch’s findings on Conformity, what can be expect about conformity in the legal system?
As witness interviews are generally done one on one, it’s likely there isn’t a large normative conformity effect. However, due to witness statements being a memory test of sorts, there is likely to be less confidence in information recall anyways - so likely to be a high informative conformity effect if other witnesses have different reports
Informational Conformity
= When someone gives the wrong answer that matches the response of a group or another person, unaware they’re answering incorrectly as they assume the group (or others) are correct in their distinction
Normative Conformity
= When someone gives the wrong answer intentionally and will full awareness, to avoid rocking the boat and disagreeing with the consensus
Co-witness Discussion Paradigms
- Original Paradigm
- Participants are shown two different stimuli (eg. a different video on their own computer)
- Participants are then tested on their memory for the video event or asked to have a chat with the other witness before the test.
- Idea is that due to the stimuli being different, in the chat phase, one participant will be misinformed and integrate incorrect information to their memory trace
Drawback; if not prompted, in the chat phase, partificpants may not isolate the point of difference from their stimulus videos. Alternatively, participants may isolate the intention of the study (realising they watched different videos) - MORI Technique
- Participants are shown the same screen, however, watch a stimulus video wearing sunglasses that are polarised in different directions
- Some participants chat before a memory test and others proceed to the test straight away
Critique; participants are less likely to identify the cause of the experiment with this method but may not realise they saw something different to the other participant - Confederate Paradigm
- One person is a confederate and the other a true particiapnt
- the confederate plans incorrect information (so can be sure point of difference is discussed
- Only one datapoint can be gained at one time however, so data collection is slower in this method
Original Co-witness Paradigm
- Original Paradigm
- Participants are shown two different stimuli (eg. a different video on their own computer)
- Participants are then tested on their memory for the video event or asked to have a chat with the other witness before the test.
- Idea is that due to the stimuli being different, in the chat phase, one participant will be misinformed and integrate incorrect information to their memory trace
Drawback; if not prompted, in the chat phase, partificpants may not isolate the point of difference from their stimulus videos. Alternatively, participants may isolate the intention of the study (realising they watched different videos)
MORI Technique
- MORI Technique
- Participants are shown the same screen, however, watch a stimulus video wearing sunglasses that are polarised in different directions
- Some participants chat before a memory test and others proceed to the test straight away
Critique; participants are less likely to identify the cause of the experiment with this method but may not realise they saw something different to the other participant
Confederate Paradigm
- Confederate Paradigm
- One person is a confederate and the other a true particiapnt
- the confederate plans incorrect information (so can be sure point of difference is discussed
- Only one datapoint can be gained at one time however, so data collection is slower in this method
Co-Witness Discussions; Gabbert, Memon and Allan 2003
Pairs of participants were shown a video of a simulated crime (video had the same content, however, differed as it was filmed from 2 different perspectives).
Unbeknownst to participants they each watched a slightly different video
Half the participants/pairs were allowed to discuss the video will eachother before being interviewed
- Age did not alter their reports on misled itemts
- 71% of participants in the discussion condition reported items they had not seen
- None of the participants in the control condition reported unseen items
- 60% of participants who had not seen the theft declared the girl guilty of stealing, after discussing the video with a co-witness (who did see the theft)
Retelling and Recounting as a source of misinformation
Illusory Truth Effect; Even the simple act of repeating a statement can increase the strength of one’s belief in its truth (makes you more confident) - this logic is why manifestation works
Audience tuning; When we recall a story, we will bias the story to accomodate the group we’re with. We may or may not have awareness of this, however, are inadvertently altering our memory by doing this
So, even in the absence of input from a conversational partner, people asked to recall a witnessed event;
1. Bias their description to reflect their own goals
2. Subsequently report believing these biased descriptions
3. Falsely recall the original stimuli as consistent with their biased descriptions
Illusory Truth Effect
Illusory Truth Effect; Even the simple act of repeating a statement can increase the strength of one’s belief in its truth (makes you more confident) - this logic is why manifestation works
Audience Tuning
Audience tuning; When we recall a story, we will bias the story to accomodate the group we’re with. We may or may not have awareness of this, however, are inadvertently altering our memory by doing this
Allport 1947 Retelling and Recounting (aka DIY Misinformation)
Explored how top down influences and stereotypes can alter our memory;
- Showed participants photos of a shabbily-dressed white man, holding a razor arguing with an unarmed, well-dressed black man
- In subsequent retellings, over half the participants transposed the characters (so it’s the black guy threatening the white guy)
Mechanisms for Misinformation
This asks how memory distortion in fact happens. There’s many possibilities but these are theories;
- The person merely reports the misinformation, but knows that it’s wrong (like normative conformity or demand characteristics)
- The original memory gets overwritten or changed
- There are now two memories, but the original one becomes less accessible (inaccessible) or the person can’t distinguish between the less
- There was no original memory (no encoding)
Toland 1990 ; Mechanisms for Misinformation
- Employed a standard misinformation procedure but participants were asked to bet money on their answers.
- The hope was the promise of money would over-rule and normative conformity or demand characteristics.
- Even with money on the table, a robust misinfo effect occured and the betting behaviour suggested there was a high level of belief in the false accusations
McCloskey and Zaragoza (1985) Mechanism of Misinformation Study
- Implemented an altered standard misinfo procedure to assess whether memory for the original stimulus is overwritten (option 2) or if the misinfo results in 2 memories with varying strengths (option 3)
- In this paradigm, participants were shown a stimulus, then a misinformation stimulus and tested. However, in the test, the misinformation stimulus was not presented. Instead, it was the original and a novel stimulus shown.
If the original stimulus memory trace has been overwritten, results should show a 50;50 original and novel stimulus response rate. If the original stimulus trace still exists, we’d expect this response to be given more often than the novel stimulus - Findings are inconclusive, there seems to be more support for option 3 (2 memory traces) however, this isn’t fully substantiated
Zajac, Dickson, Munn and O’Neill 2016 ; Mechanisms of Misinfo study
Used a confederate misinfo paradigm and convinced participants the confederate was drunnk (with comprehensive methodology).
Then, the pair watched a video of a theft and had a discussion before an interview/memory test.
The confederate (who was not in fact drunk) gained a pre-misinformation response (eg .by asking what colour was the car?) before introducing two pieces of misinformation (eg. i think the car was red, not blue!).
- When pre-misinformation responses were discrepant with the misinformation, the participant was more likely to accep misinfo from a sober co-witness than the drunk one
- When participants weren’t able to provide a pre-misinformation response (ie. they couldn’t remmeber the colour of the car) the co-witness condition had no effect on the misinformation acceptance (drunk or not, participant would accept the misinfo at a standard rate)
Assefi and Garry 2003 ; Misinformation Mechanism Study
- Participants drank a tonic water beverage, but half the group were told it was alcohol
- Participants then took part in a standard misinfo paradigm
- findings saw that participants in the ‘alcohol’ group were more susceptible to misleading post event information, however, encoding did not differ between the groups (as in the control condition when not giving misleading post-event information, ‘drunk’ participants did just as well as sober ones).
Implications For eyewitnesses; Memory Storage
What happens between the event and the memory interview is of the utmost importance
- eyewitness evidence should be treated as we treat physical evidence (collect it quickly and do not contaminate it)
- disortion can be minimised but not avoided
