2007-09-17: The reliability of eyewitness testimonies and of confessions 19
Part 1: Reliability of eyewitness testimonies 19
Estimator variables 19
System variables 20
What about eyewitness testimonies? 21
What do eyewitnesses forget? 24
Co-witness discussions 24
Social contagion of memory (Roediger, Meade, & Bergman, 2003) 25
Memory conformity paradigm (Gabbert et al., 2003, 2007) 25
Age differences in children (Candel, Memon, & Al-Harazi, in press) 26
The reliability of confessions 27
False confessions in the lab 28
Kassin & Kiechel (1996): The computer crash paradigm 28
New method: false video evidence (Nash & Wade, under review) 29
New method: False video evidence 29
Literature suggestions 31
2007-09-25: Remembering Childhood Trauma 32
Long term memory for child sexual abuse (CSA) 32
Theoretical reasons 32
National Institute of Justice Study: Emotional Effects of Testifying on Child Sexual Abuse VictimsGoodman, Taub, Jones, Rudy, England, Port, & Prado (1992) 33
Research design 34
Main findings 34
National Science Foundation Study: Long-Term Follow-Up 35
Of theoretical interest: What failed to predict non-disclosure? 37
Summary: Lost memory / nondisclosure of abuse 37
Study 2: Can people recover lost memory of abuse 38
Research question 38
Regression analysis: Can we predict subjective forgetting? 39
Reports of reasons for forgetting (N=21) 40
Do reasons for subjective forgetting relate to abuse characteristics? 40
Descriptions of stories of people who had a time when they couldn't remember 40
Remembering Childhood trauma part II 43
Study 3: Long-Term Memory for Child Sexual Abuse 43
PTSD: Definition and rates 43
Memory coding 45
Predictors of CSA memory 45
Why the interaction effect? 46
Children’s Memory for Traumatic Events 47
Memory for negative / stressful events 48
Socio-emotional factors 48
Childhood amnesia 48
Frequent research paradigm 49
Conflicting findings 51
Voiding Cystourethrogram Fluoroscopy (VCUG) 51
Bowlby attachment theory and children’s memory 53
Disorganized “D” attachment 54
Back to the VCUG procedure 55
Individual differences in trauma memory: Attachment theory 57
Memory in adults with CSA histories: Individual differences 57
Eisen 1998 58
False memories 61
2007-10-08: Remembering names 63
“Tip-of-the-tongue state” (TOT) 63
Characteristics of TOTs (since the 1960s) 63
Now on to learning new people 65
Another type of TOT (Schwartz' illusory TOTs) 66
Does everyone have TOTs? 66
Why are names hard to remember? 67
Jacoby’s “false fame” effect 68
Young, Hellawell & Hay (1985) 68
The Thatcher illusion (Thomson, 1980) 69
2007-10-15: Remembering names after brain injury 70
The point 72
Proper name aphasias: (egennavn) 73
Dom Thomson (Australian researcher) 75
The exam 76
2007-09-14: Children and false memories
Ingrid Candel, Maastricht University
Today’s topic is children and false memories. There is some overlap with the last lecture (Tim’s) but this one is focusing on children
Is it is possible that this child is claiming to have experienced an event that did not happen? Today I’ll show you that this can happen.
McMartin preschool chase: well known in the USA. In that case 7 teachers were accused by hundreds of children for sexual abuse. These researchers, McGoven and colleges analyzed the investigative interviews, and it appeared that suggestive techniques were used which have might led to false memories. This has stimulated the study of false memories in children.
Bidrose and Goodman study (2000)
In studies on real victims it is hard to know what happened to them, and if we want to study if memory is false or not we have to be sure what happened. In the Bidrose and Goodman study they knew what happened. They did a case study on 4 girls, where there was evidence of the abuse (photographs and audio recordings), so they could compare the facts with the testimonies. They found that for almost 80% of the allegations there were some evidence in pictures or audiotape. For about 40% of the acts they didn’t give testimony about them. There was no evidence for about 20% of the allegations. One could argue that the testimonies about these acts might be false memories. Studies like this one are very rare, since we rarely know what happened.
So, we have to find another way, using lab studies:
False memories for details and false memories for whole events should be kept apart. The papers in the classfronter (Kimberly &..) are about this topic.
You can imagine that you add details to an event, is it a false memory?
You can also study false memories for entire events.
The paradigms for studying this are different
False memories for events
The DRM (Deese-Roediger-McDermott) paradigm, uses the critical lure mechanism. You study a list of words all related to a more central word which is omitted. E.g. sleep. One might argue that this is a false memory for detail, not the entire list. Adults are likely to falsely report remembering this word when asked if the word was in the list. What about children? There is a difference. One of the first studies, Mark Haus’ study in 1995:
Compare the light blue bars from the graph. Adults (b) report more false memories than children(d), so they are more susceptible. A reason for this might be semantic networks. The DRM paradigm activates semantic memory, and we make source-monitoring errors. Adult semantic networks are more extensive than children’s semantic memories, so it is more likely that adults have more errors. Some researches claim that the DRM paradigm has low ecological validity, as it is not natural, and completely different from false claims about sexual abuse. To solve this problem researches started to use emotionally negative lists instead of neutral ones. What happens if they study negative word lists? Do they report the same number of errors?
Howe (2007) studies this with children:
3 neutral DRM-lists: chair, fruit, sweet
3 negative-emotional DRM-lists: anger, cry, lie
Children were more likely to develop a false memory for neutral words as compared to emotional ones. He also found an effect of age: 12 year olds are more likely to develop false memories as compared to younger children. The finding that neutral word lists elicit more errors than emotional studies is quite robust, and have been replicated in Maastricht. In this study they used 5 neutral and 5 emotional lists:
Still, one could argue that this paradigm is not that ecologically valid, it is might be more valid to look at the performance of PTSD patients. How do these people, who have experienced a traumatic event, perform on a DRM task? E.g. Tim’s study on Bosnian PTSD patients. They recall more critical lures than control patients, so they’re more likely to develop false memories than non-PTSD patients, and they’ve got problems distinguishing between internally activated and externally activated memories.
PTSD – posttraumatic stress disorder
After traumatic experience
Reexperiencing the trauma: flashback and nightmares
10 trauma: blood, concentration camp, war, shell, tears, funeral, wounded, Sarajevo, rape, refugees
Significant Group x List interaction, F(1,98) = 7.34, p < .01; PTSD-patients falsely recalled more war-related critical lures than did control participants. might be beacause of source monitoring errors.
So far: Children are less likely to recall non-presented words. Here we see a developmental trend. Children falsely recall fewer emotional negative words than neutral words. PTDS increases the recall of the trauma related non-presented words (in adults).
Comments: It could be because of the stronger relations between the nodes might activated faster, the critical word must be activated, then you make an error, and in PTSD patients it might be faster.
The Misinformation Paradigm
The idea is that a lot of things happen in between the experience of an event and the moment you have to recall that event. The event and the recall of that event don’t have to happen on the same day. You can read about the event in the media, you might have talked with other victims or friends and family about the event, and as a result you might encounter misinformation. What is the effect of this information on your later recall? It might be that children in particular develop false memories as a result of the misinformation.
Studies started with Loftus and Palmer in 1974: They wanted to know what the effect of misinformation was on adult’s memories. They showed participants with a video of a car crash. Then they asked how fast the car was going when they hit each other or “.. smashed into each other”. The last question suggests that they were going much faster. The participants where then asked to give an estimation about how fast they thought the car was going. If prompted with the “smashed variant” it elicited a much higher answer. Then, they had to answer “did you see any broken glass”. The word smash led to much higher degree of false broken glass memories.
Verb Mean speed estimated
So, a suggestion: Misinformation clearly has an effect on the development of false memories. What is the effect of misinformation on children? Sutherland & Hayne 2001:
24 children, M = 11.59 years
24 adults, M = 20.21 years
Video about child who becomes separated from caregiver during shopping trip
Neutral: “In the video Mary was given a bear. Who gave her the bear?”
Leading (consistent with content): “In the video, Mary was given a white bear. Who gave her the bear?”
Misleading (inconsistent with content): “In the video, Mary was given a green bear. Who gave her the bear?”
Memory test: “What color was the bear?”
As you can see, both children and adults are more likely to report a false memory in the misinformation category as compared to the neutral and leading questions. Important: Remember, children are more likely to develop a false memory than adults. In the DRM pattern the reverse is true (NB!). In summary: Using a DRM paradigm the general finding is that younger children develop less false memories than older children, in the misinformation paradigm the reverse is true. This can be explained by semantic networks for words. In DRM younger children have less extensive networks compared to adults, and are therefore less likely to develop false memories as compared to adults. In the misinformation paradigm you test your level of suggestibility, children are more likely to accept misinformation compared to adults. Children are more likely to accept what adults suggest, on the other hand children’s memory is poorer when compared to adult memory, there are holes in their memory, and suggestions fill in holes. The misinformation and the DRM paradigm tests different mechanisms (semantic network vs memory)
It is not that easy to use lab results in a court, you can not say that children develop fewer false memories than adults, so for that reason the children in the McPearson case might all have been right, it is much more complicated.
Summary: Until now, many studies have shown that children easier develop false memories, and they change memories. Only a handful of studies have focused on omission errors, are children likely to omit information based on a suggestion?
This is an important question, because clinical and forensic interviews ask children questions about things they know little about, this makes it possible that the questions suggest that things might not have happened actually happened and vice versa. There are 3 types of suggestions:
False memories for details (commission errors)
The design is a mixed design (M study, 38 7-year-olds and 47 11-year-olds): All children take part in a target event, a presentation about China. The presenter tells the children, showed pictures and so on. After three days the children were interviewed separately. During the questions different types of questions were asked:
After this they were involved in a free recall task. The presenter was not wearing glasses.
Commision question: “The presenter was wearing glasses, wasn’t she?”
The recognition question: “The presenter was wearing glasses”. First of all, we scored the children’s correct recall. Older children had a better memory than younger children. Children were more likely to accept information involving a change (red vs green) as compared to commission and omission. Younger children are more susceptible to all three than compared to older children.
1 point for each correctly recalled detial (inter-rater-agreement (r) = .97)
Molder children = 17.09, SD = 4.57
Myounger children = 11.26, SD = 3.74
t(82) = -6.30, p < .01
In conclusion: Younger children are more suggestive than older children. This whole finding is consistent with the discrepancy detection principle. “When I see nothing, and you suggest that you see a red car”, it is more likely to distrust the suggest, but when you see a car there is at least some overlap, and you are more likely to accept the suggestion.
Back to the McMartin pre-school case
Children are likely to report false details about or to change the details of experienced event.
But are they also likely to report entire false events?
Children are more likely to report false details, but are they more likely to report false information about an event they have not experienced? There is only one valid paradigm for addressing this issue, the implantation method, using suggestion:
You can present them with suggestive or false narratives
You can present them with suggestive or false photographs
You try to implant memories for not-experienced events. the first to use this paradigm were Loftus & Pickrell (1995).
Loftus & Pickrell (1995)
24 adult participants
3 true narratives, 1 false narrative (lost in a shopping mall)
3 test occasions
“You, your mom, Tien, and Tuan all went to the Bremerton K-Mart. You must have been 5 years old at the time. Your mom gave each of you some money to get a blueberry Icee. You ran ahead to get into the line first, and somehow lost your way in the store. Tien found you crying to an elderly Chinese woman. You three then went together to get an Ice cream”
T1: 7 (29%)
T2: 6 (25%)
T3: 6 (25%)
The participants listened to 3 true narratives and 1 false narrative (lost in a shopping mall). The parents of the participants helped in the experiment. They interviewed the participants on three test locations, “tell me as much as possible about these events”. Are they going to give information about being lost in a shopping mall, something which did not happen? The experimenter included some personal information to make it more likely. The instruction was to “tell me more, give me as much information.
This lead the researchers to believe that people can be led to believe entire events that did not happen. Both adults and children are likely to develop false memories about such events.
Quick repetition: The false photographs: 20 participants where tested, three true photographs and one false one. The participants were instructed to give as much information about the photographs as possible. After viewing him or herself in the hot air balloon this was the report:
“…Um basically for $10 or something you could go up in a hot air balloon and go up about 20 odd meter... it would have been on a Saturday and I think we went with, yeah, parents and, no it wasn’t, not my grandmother…not certain who any of the other people are. Um, and I’m pretty certain that mum is down on the ground taking a photo”.
People reported quite detailed false memories, such as who took the photograph. On the second interview after 1 week 60% of the participants had some kind of false memory. This is a quite dangerous that it is so easy to implant a false memory in participants. After this study using the doctored photograph paradigm, other researchers have used this paradigm, and both adults and children are susceptible:
Surviving an animal attack (Porter, Yuille, & Lehman, 1999)
Winning a contest (Ost, Foster, Costall, & Bull, 2005)
Having an eventful birthday (Hyman, Husband, & Billings, 1995)
Receiving a rectal enema (Pezdek & Hodge,1999)
Lindsay, Hagen, Read, Wade, & Garry, 2004
Both adults and children are likely to develop false memories as a result of the presentation of false photographs, for a variety of events. An argument is that people are unlikely to meet these kinds of events. Lindsay et al wanted to check if true photographs elicit false memories, when they listen to a false narrative.
Adult participants listen to a false narrative with and without a true photograph (of their class). The situation they were asked about was putting slime in a teacher’s desk. Participants were interviewed twice about this false event. Participants in the photograph condition were more likely to develop false memories.
In summary: Adults and children can develop memories for events that did not happen, but at least one can argue that “lost in a shopping mall” is something completely different than being abused. It is a completely different type of event, being lost in a shopping mall is much more plausible for a child to happen than being sexually abused. This stimulated researches to study the factor of plausibility in false memories.
Plausibility in false memories
Pezdek & Hodge study from 1999: They created a plausible false event, the lost in a shopping event, and implausible: receiving a rectal enema.
Results (38 children):
54% no false memory
14 children “remembered” plausible but not implausible event
“It should be easier to plant false memories of childhood sexual abuse with children for whom childhood sexual contact with an adult is more plausible than with children for whom childhood sexual contact with an adult is less plausible.”
Only one child remembered the implausible but not the plausible. Researches concluded that it is much easier to implant the memory of sexual abuse if it is plausible, if they’ve received physical abuse e.g.
In the Strange, Sutherland, & Garry (2006): Suggestive photographs study two scenarios were used:
Plausible false event: hot air balloon ride
Implausible false event: drinking a cup of tea with Prince Charles
They found that children were equally likely for both. Plausibility is not a factor.
However, nobody so far have studied the role of valence, and that might be one of the reasons for the different results in the two above. Misinformation, plausibility and valence are mixed together.
Study 1: False narratives
68 children, 8.65 years (SD = 0.67)
3 true narrative
1 false narrative:
Implausible/positive: “being the leading actor in a movie”
implausible/negative: “attacked by a monkey”
plausible/ positive: “being on the top of the Eiffel Tower”
plausible/negative: “almost choked on a candy”
The plausibility were deiced based on a pilot study, where kids rated the plausibility of 16 events.
Children were interviewed twice, with one week between each interview, about 3 true and one false narrative. Children were instructed to give as much information as possible about the events. The result: Valence didn’t play a role for improbably events. They were more likely to have a false memory about “choking on candy” than “being on top of the Eifel tower. Although, they controlled for general plausibility one might argue that the two differ in personal plausibility and script knowledge. It might be that children are more likely to develop a false memory about choked on a candy, and they might have more script knowledge – they might not know about the Eifel tower, but have seen a friend choke on a candy, etc. The same applies to the rectal enema, it might be that children don’t know about this situation, they don’t know what it is and can’t develop a false memory. For that reason there was a second study, with a false photo study, where all three possibilities were tested:
These events only differ when it comes to valence (tested with children).
This was the worst photograph we could show them, we could not show them photographs of the participants undergoing an operation or something like that.
Valence plays a role in the development of children’s false memories for plausible events.
“almost choked” and “Eiffel tower” might differ in terms of personal plausibility and script knowledge
“General plausibility does not imply personal plausibility” (Scoboria, Mazzoni, Kirsch, & Relyea, 2005, pp. 793).
Prior knowledge affects the development of false memories (Pezdek, Finger, & Hodge, 1997; Pezdek & Hodge, 1999)
Example of false memory:
Child: This was when I was in the hospital in Hasselt. I was about 3 or 4 years old. And my hand was in plaster. And then a few people came to visit me and mum and dad were the whole day with me. Sometimes it was hard for me to sleep…um….I also got a present from mum and dad.
Interviewer: How did it happen?
Child: Because I dropped a glazed bottle and I wanted to pick up the pieces and then I cut myself badly. And then my hand was broken and I had to go to the hospital.
The story is quite detailed.
When one controls for general plausibility and script knowledge: These two are very important for the development of false memories, valence doesn’t play any role when controlling for these two.
Number of children who developed a false memory for the hot air balloon ride (positive event) = number of children who developed a false memory for the admission into hospital (negative event)
Conclusion: When controlled for general and personal plausibility, and script knowledge, children are equally likely to develop false memories for positive and negative event.
How can you be sure that these children developed a false memory? Maybe they wanted to please you and came up with a story? On the end of the interview the kids were debriefed “I made it up, it never happened to you”. The response of the children is that the children indeed developed a false memory, “I really remember it, I really think the photograph is real”, it is not only compliance, it also has to do with memory.
True photograph method (Lindsay et al. 2004)
Comparing development of false memories for negative and positive events children are just as likely to develop false memories, so it could be that both tap the same mechanisms. What would happen if we compare false memories for a negative and a neutral event? This was tested by using the true photographs paradigm:
A sample of children rated about 20 events on a scale.
Example of false memory report for copying off your neighbor.
It is quite detailed, the child reports details not in the false narrative, the child elaborates, indicates his or her emotional state as this happens. We only found a significant effect of valence, there were more false memories for the negative events as compared to the neutral one. The photograph had no importance. It could be that children are more acceptable, they don’t need a photograph, just a narrative is enough. There might also be a kind of ceiling effect, there is not a lot of room for the photograph to elicit false memories.
Non-significant Valence X Photograph interaction
Non-significant main effect of photograph
Significant main effect of valence at both interviews; the negative event elicited more FM than the neutral event (ps < .05)
This is important because what we would like to do is to apply our lab findings to the real world. Now we have more support for that children are more likely to develop a false memory for a negative event. All the negative information is more integrated in our memory, as soon as an aspect, a detail of that information is cued, like with a photograph, that might result in the activation of other negative information. This activation might not apply to neutral information, which are not as integrated in memory.
Being abducted by a UFO
The false events I’ve talked about so far, copying off your neighbor, moving to another classroom, etc. there is a huge difference between these and sexual abuse. What about highly implausible events? Are children more likely to develop those? The rectal enema has no script knowledge, but how about being abducted by UFOs? According to Mazzoni et.al you have three factors which are needed:
Three-step model of false memory formation (Mazzoni, Loftus, & Kirsch, 2001):
Evaluate event as plausible
Belief that event did happen
Interprete images and thoughts as memory details
If you believe that the event happened to you might interpret the images and thoughts as memory, and if that happens you might start to develop a false memory. What we did in a study, we presented kids with prevalence information to make it more plausible that this happens (information about the frequency of the event falsely indicating that the event happens more often that what you think makes us more likely to believe that it has happened to you on a self-rating. After reading false prevalence information participants are more likely to believe that an event has happened to you). You might argue that beliefs and memories are different, but so far only beliefs have been studied. Previous studies: prevalence information increases the belief that a false event has happened (e.g., Hart & Schooler, 2006; Mazzoni, Loftus, & Kirsch, 2001)
Almost choked on a candy or abducted by UFOs. The information was that when kids were 4 year old it happened very often that UFos were shown or “candies were taken out of the supermarket because they were dangerous”.
Almost choked on a candy: “your mother told me that you were at a birthday party when you were 4 years old. At this party you received a bag of candies. When you were at home again, you were allowed to have one cany. Your mother saw that you turned blue Then she hit you on the back and the candy came out”
Abducted by a UFO: “Your mother told me that when you were 4 years old, you were abducted by a UFO. This happened when you were alone outside. Your mother was inside the house. Then she suddenly saw through the window that a UFO took you”
Prevalence information: false newspaper article
Children were assigned to one of the 4 categories. Design: 2 (age: younger children vs. older children) x 2 (false event: plausible vs. implausible) x 2 (prevalence information: yes vs. no)
They were instructed to give as much (false) information as possible about the event. There is a prevalence interaction, the prevalence information is something young children are much more susceptible to. Plausibility was not a factor. There was one factor so far (earlier) about plausibility. So far, two studies show an effect, one study show no effect. What is interesting in this study is that almost a third of children have a false memory about a very obvious false memory. Again, the reports are quite detailed, “scary, emotionally negative”, so if children are more likely to have false memories for such negative implausible events there might be a parallel with abuse memories.
Prevalence information might also be given to children in the form of testimony of other children, like if kids at school start to talk about sexual abuse, if one kid talks about sexual abuse a kid might get the idea that it is much more prevalent, this might also apply to the McMartin preschool case.
Example of false memory report UFO abduction:
Child: “I saw cameras and flashes and some persons in the UFO.”
Interviewer: “How many persons did you see?”
Child: ”approximately nine or ten.”
Interviewer: “What kind of persons?”
Child: “Persons like me, children.”
Interviewer: “What else did you see?”
Child: “I saw some persons and also some blue/green puppets were passing
Summary: Bad news, children are likely to report false details and even false events, all kinds of events. The good news is that without exposure to suggestive information children are highly accurate. This comes from interviews with children of a very negative event that they had to experience, and they stay very accurate after a couple of year. The problem with children is that they are not very eager to talk about what happened to them, especially, in an investigative interview, they are often very quiet, the police has to get information, and might use suggestive questions, or imaginative techniques, information from other eyewitnesses and so in, this increases the chance of false memories.
Garven, S., Wood, J., Malpass, R., & Shaw, III, J. (1998). More than suggestion: The effect of interviewing techniques from the McMartin Preschool case. Journal of Applied Psychology, 83, 347-359.
Howe, M.L. (2005). Children (but not adults) can inhibit false memories. Psychological Science, 16, 927-931.
Loftus,E.F., & Pickrell, J.E. (1995). The formation of false memories. Psychiatric Annals, 25, 720-725.
Pezdek, K., & Hodge, D. (1999). Planting false childhood memories: The role of event plausibility. Child Development, 70, 887-895.
Strange, D., Sutherland, R., & Garry, M. (2006). Event plausibility does not determine children’s false memories. Memory, 14, 937-951.
Wade, K.A., Sharman, S.J., Garry, M., Memon, A., Merckelbach, H., & Loftus, E. (2007). False claims about false memories. Consciousness and Cognition, 16, 18-28.
2007-09-17: The reliability of eyewitness testimonies and of confessions
Ingrid Candel, Maastricht University
As you know evidence can come from different sources, such as witnesses, and suspects in form of confessions.
Today I’ll talk about the reliably of eyewitness testimonies, the second part about confessions.