Pozzulo et Al CA Flashcards
The psychology being investigated-Eyewitness Testimony
People present at scene of an accident/crime are often questioned by police and asked to provide a detailed description of what they saw, heard, or smelled. This is called an eyewitness testimony.
-Important part of many police investigations + often provides crucial evidence in subsequent legal proceedings.
-Although jurors often find eyewitness testimonies compelling, decades of research suggest that they are unreliable. Memory is reconstructive, meaning it may be altered with each retelling + can become contaminated by post-event info.
-memory also degrades over time, underlining the importance of developing techs that allow officers to elicit accurate info as quickly as possible.
-Hence, eyewitness testimonies have become a rich area for forensic researchers and their work has directly impacted the way witnesses are interviewed.
The psychology being investigated-Eyewitness testimony 2
-One problem however, is that memory distortion can lead people to remember things that didn’t actually occur-or forget ones that did.
-Research suggests that people often “fill the gaps” in their memories in a process called confabulation.
-Overtime, these inaccuracies can become indistinguishable from the more accurate aspects of memory.
The psychology being investigated-False positive responses
Forensic researchers often ask pps to observe a staged crime/event and then identify the culprit from the selection of people-called a line up.
-Sometimes culprit is present (target-present line ups), and sometimes not (target-absent line ups).
-There are a number of possible responses in a target-present line-up: a correct identification, a false positive response (incorrectly choosing a foil), or a false negative (incorrectly stating that the suspect is absent).
-In a target-absent line up, pps are correct is they reject the line-up(no possibility of false negatives), but they still make a false positive response by selecting a foil.
Background-General
The reliability of children as eyewitnesses is relatively understudied compared to adults, but Canadian researcher Dr Joana Pozzulo dedicated her research to reducing wrongful convictions through the development of evidence-based, age appropriate techniques for working with child-witnesses.
Background-Line Ups and Identification
-Pozzulo et Al’s research focuses specifically on use of line-ups in identification of suspects.
-Foils are generally people who look physically similar to the suspect.
-A variety of line-up formats exist, including live line ups (physically present, seen through 1-way glass, and photographs and/or videos of suspects).
-Witnesses may view individuals simultaneously or sequentially.
Background-Children as witnesses
-Joanna Pozullo + Roderick Linsay conducted a meta-analysis, demonstrating that children as young as 5 are relatively accurate when faced with a target present, simultaneous line up, but children as old as 13 struggle to correctly reject target-absent line-ups.
-While these errors are likely due to a combination of cognitive + social factors, Pozzulo et Al. highlight the role of social influence and desire to please interviewers (social d. bias) as prominent issues affecting their accuracy with target -absent line-ups.
-As children are more likely to identify a foil, not realizing that rejecting line-up is also an accepetable response-
Pozzulo et Al. In brief
-To explore whether children’s errors in line-ups result more from social or cognitive factors, Pozzulo et Al. manipulated the familiarity of the target.
-They suggested that children who can pick out a popular cartoon character, with 100% accuracy, logically should be able to reject a target-absent line-up, where the character isn’t present.
-If they do not do this, social factors are implicated as a case of false positive responses when identifying human faces.
-In their study, 4-7 year olds and adults where shown films of easily recognizable cartoon characters and unknown human actors, before being presented with either a target-present/target-absent line up.
-As expected, children were easily able to identify cartoons in target-present line-ups but still made mistakes on target-absent ones. They made more mistakes than adults and this was even more pronounced with unknown human faces compared with cartoons.
-The influence of social factors appears to be at least partly responsible for children’s errors, suggesting that minimizing these factors could allow children to give accurate testimonies.
Aims
Study aimed to explore the role of social and cognitive factors in children’s identification of target faces in line-ups. Specifically, they aimed to investigate whether children:
-are less able to recognize human faces than adults.
-make more false positive identifications than adults when faced with:
–target absent and present line ups
–human and cartoon faces
Research Methodology 1
Pps observed video clips in a controlled setting before trying to identify the target person from the video in different types of line-up. 2/3 IVs were manipulated by the experimenter, making it a lab experiment.
Research Methodology 2
Design+Variables:
3 IVS:
-Age: young children/adults
-nature of target faces (familiar cartoon characters or unfamiliar faces)
-type of line-up target-present or target-absent.
The first IV compares data from diff people (IVs),while the 2nd and 3rd IVs compare data from diff conditions, involving the same people (repeated measures design).
The 4 vids (male cartoon-Diego and Female-Dora, male actor, female actor) were randomized for each pp to avoid order effects. The order of the faces (targets and/or foils) in line-ups also randomised.
DV: The recognition rates of children and adults when identifying faces
Sample
Children aged 4-7 (21f, 38m mean age= 4.98 years, SD=0.82), selected from 3 private schools in eastern Ontario, Canada.
Adults aged 17-30 (36 females,17m, mean age=20.54, SD=3.34), selected from introductory psychology pp pool of an eastern Ontario University.
Procedure 1
Researcher created 4 video clips, 2 starring the familiar cartoon characters Dora and Diego. Dora was speaking to the audience in one video and Diego was putting on safety gloves in the other.
The other 2 clips separately included a male + female main character, but human this time, matched for age and ethnicity-both were 22 and Caucasian.
-The woman was brushing her hair and the man putting on his coat. All clips were in color, no sound, 6 secs long, featured a 2-3 second close up of a person’s face. There was only ever 1 character per clip.
Procedure 2
Each Video had an accompanying photo array line up of 4 tightly cropped, black and white headshots.
-In the target present line ups, one of the photographs was the target person/cartoon character (but dressed differently from video clip) and the other 3 photographs were foils, meaning people who looked similar (e.g similar facial structure, hair length and color), but weren’t target.
-In the target absent condition, target was swapped for another foil in the same position. All line ups included a silhouette as a visual reminder that genuine target may not be present. All images viewed together, at same time (simultaneously).
-Human foild were selected from a set of 90male and 90female faces by a panel of 3 researchers.
-Cartoon foils were selected due to similarity to target. As Dora + Diego generally wear the same brightly colored clothes all the time, images for line ups were tightly cropped and black and white photographs. This meant pps couldn’t identify them based on clothing-rather than faces.
Procedure 3- Before the testing phase
The parents of the child pps completed informed consent forms and the Demographic and caroon watching form, an 8q questionnaire.
-Parents asked to estimate how long their child spent watching cartoons per week and how much watching Dora the explorer and Go Diego Go! Also asked child’s age, gender, primary language, and ethnicity and whether they had any siblings.
Children tested individually at schools. The 4 female researchers were introduced as people from the Uni doing research on tv shows + computer games.
Before starting the experiment, the children and researchers participated in some craft activities together to get to know each other+get comfortable.
The adult pps completed their own paperwork + were told the study was about memory.
-now there was clearly some deceitefulness present, and the fact that adults and children were told diff things (memory vs computer games might have influenced how they acted (e.g adults were on guard mrmorising vs children weren’t so it wasn’t fair).
Procedure 4-Watching the Videos
The child pps told that they would be watching some videos and to pay attention as they would be asked some questions and shown some pictures afterwards.
-The first video was shown once the child seemed happy + ready.
-After each clip, they were asked “What did the cartoon character/ person look like?” and “Do you remember anything else?”-if they did not answer, asked: “Do you remember anything
from the video?”
-the researcher spent 2mins on this filler task and wrote the children’s answers down, although they would not be analyzed. The procedure was identical for the adults, except they wrote down their answers.
Procedure 5-The Photo-array line-ups + after testing was complete
Pps given the following standardized instructions : “Please look at the photos. The person/cartoon from the video may or may not be here. If you see the person/cartoon please point to their photo. IF not, please point to this box.
The line-up shown using presentation software on a laptop. The adults completed the same procedure, but recorded answers on a sheet rather than pointing to a screen.
Results-target-present line-ups
As expected, the children were significantly better at identifying familiar cartoon characters in target present line ups than unfamiliar human faces. Their success rate was 0.99 compared with just 0.23 for unfamiliar human faces.
The adults were also significantly better when face with target present cartoon line ups (0.95) than unfamiliar faces, but their success rate was much higher than children’s (0.66). There was no significant difference in the success rate of adults and children in identifying cartoon characters.
Results-Target-absent line-ups
Again, as expected, the children had a significantly higher success rate with target absent line ups including cartoon characters (0.74) compared with human faces (0.45).
The adults also performed significantly better with the cartoon characters (0.94) compared with human faces (0.70), and their performance was also significantly better than children’s for both conditions.
-children had a much higher rate of false positives (incorrectly choosing a foil) compared with adults. Although the rate of false negatives (incorrectly stating that target is absent) was also higher for children, the diff was smaller than it was for false positives.
As the children were clearly able to pick out Dora and Diego in target present line ups, the lower rate of correct responses (0.74) in the target absent ones demonstrates that their false positives are likely to be driven by social factors rather than cognitive ones-that is, the children probably knew the character wasn’t there, but felt that the social situation required them to pick someone out, despite being told this wasn’t the case.
Results Data
Response rates for identifications from target-present and target-absent line-ups. (False negative = participant incorrectly rejected the line-up when target was present.
False positive = participant incorrectly identified a foil as the target.):
-Target present
—Children
Dora: 1.0
Diego: 0.97
—-False negatives: 0.03
—Adults
Dora: 1.0
Diego: 0.89
—-False negatives: 0.11
-Target absent
—Children
Dora: 0.80
—-False positives: 0.20
Diego: 0.67
—-False positives: 0.33
—Adults
Dora: 0.96
—-False positives: 0.04
Diego: 0.92
—-False positives: 0.08
Human faces
-Target present
—Children
Female: 0.24
—-False positives: 0.38
—-False negatives: 0.38
Male: 0.21
—-False positives: 0.45
—-False negatives: 0.34
—Adults
Female: 0.46
—–False negatives: 0.54
Male: 0.85
—-False positives: 0.15
-Target absent
—Children
Female: 0.47
—-False positives: 0.53
Male: 0.43
—-False positives: 0.57
—Adults
Female: 0.72
—-False positives: 0.28
Male: 0.67
—-False positives: 0.33
Conclusions
The researchers concluded that since the children were able to identify the cartoon characters with almost 100% accuracy in target present line ups, cognitive factors (e.g faulty memory) weren’t responsible for the lower success rate in correctly rejecting the foils in the target absent line ups. They were clearly able to recognize the characters and, therefore, logically should have been able to recognize that none of the foils was Dora or Diego; yet some of the children still made errors. Pozzulo et Al concluded that these must have been caused by social factors-that is, incorrectly believing that the researcher would prefer them to make a positive identification, regardless or whether it was right or wrong, and despite the researcher saying that the target person may not be in the line up.
It was also concluded that, as expected, children are less accurate than adults when faced with unfamiliar human actors and generally more prone to give false positive responses (incorrect identification).
Evaluation-Ethical Issues +
One strength is that, despite exploring false memory and using target-absent line-ups, there was no deception. All pps were informed before the study began that they would be asked about the videos afterwards and when shown the photo-array line ups were told that person may not be present.
This is important as studies of false memory often attempt to manipulate people’s recall, which would make them vulnerable to psychological harm and hence meaning that fully informed consent is not possible.
-the children were rewarded, all pps thanked, children played with, full consent forms, encouraged/reminded to withdraw whenever if needed-without reproach or repercussions.
Evaluation-Methodological Issues-Reliability
A strength of the study is the standardized procedure + instructions, which means the study can be easily replicated. All pps were told to watch carefully as there would be questions afterwards and pictures to look at. They all completed a 2min filler task, answering a researcher’s open qs about what they remembered. Finally, they were all given the same instructions about identifying the target, including the fact that the person may or may not be present in the line up. This means the study can be easily replicated to check whether the high level of false positives responses in target absent line ups is reliable.
But a weakness regarding reliability is the fact that the researcher didn’t ask evryone the exact same questions, as part of the 2min filler task-some pps were asked 2 questions and some 3. This depended on how mcuh info they had already provided. These questions may seem inconsequential, as the data was not analysed, but pps may have gone into the line up task with different confidence levels based on how much they had answered the previous questions and this might have altered how they responded in the actual line up task (saying more aloud might have also improved memory).
Evaluation-Methodological Issues-Validity 1 +
-Strength: Use of repeated measures design, increased internal validity of findings.
-Pps took part in both types of line ups with both types of targets (human/cartoon). Performance in target- present line-up helped provide a baseline with which performance in the target-absent condition could be compared.
-If the researchers had used an Independent measures, the argument that poorer performance in the target-absent condition resulted from social not cognitive factors would not be so compelling-this is bc it could have been caused by pp variables, that is, one group might have had slightly worse or better recognition abilities or working memory skills.
Evaluation-Methodological Issues-Validity 2 -
-weakness of design: although researchers suggested that they were manipulating familiarity when using popular cartoon characters vs unknown human actors, they were actually altering 2 things: familiarity + whether the character is a 2D animation or a 3D human.
-This is important as it makes it hard to interpret findings as it is possible that we are just better at remembering 2D characters due to decreased complexity as opposed to familiarity.
