Samenvatting
Summary Psychology of Eyewitness Testimony Task 1 - 5
- Instelling
- Maastricht University (UM)
Maastricht University specialisation Legal Psychology: Summary/Samenvatting Psychology of Eyewitness Testimony
[Meer zien]Voorbeeld 4 van de 48 pagina's
Voorbeeld 4 van de 48 pagina's
In winkelwagengesponsord bericht van onze partner
Voorbeeld van de inhoud
Task 1 “Witnesses are unreliable, aren’t they?”Ackerman & Goldsmith (2008) – Control over grain size in memory reporting-with and
without satisficing knowledge.
Approaches one might use to decide what answers a person volunteers to give:
Monitoring and control processes that underlie memory performance
Social-communication approach = approach that focuses on cooperative pragmatic principles that are
involved in the explicit and implicit communication when one person asks a question to another person in a
specific social context.
The given information depends on the listeners and their needs or favours.
Metacognitive approach = approach that focuses on both cooperative pragmatic principles and monitoring
and control processes that underlie memory performance.
Granularity= precision or coarseness (of answers to memory questions)
The control of grain size in memory reporting is guided by two objectives:
Correctness
Increasing the chance that an answer is correct implies increasing the coarseness of an answer.
Informativeness
Increasing the chance that an answer is informative implies increasing the precision of an answer.
An appropriate compromise between correctness and informativeness needs to be made in order to choose a grain
size for answers.
Grain-control model (Goldsmith) = model that states that the aim of correctness is bigger than the aim of
informativeness
Satisficing model (Goldsmith) = model that states that a respondent strives to provide as much information as
possible as long as the perceived probability of correctness satisfies a reasonable minimum level. The level of this
minimum-confidence criterion is based on the emphasis on correctness.
Relative expected-utility maximising model (Goldsmith) = model that states that a respondent calculates the
subjective expected utility of possible answers at various grain size, compares these values and chooses the answer
with a maximal subjective expected utility.
Dual-criterion satisficing model (Ackerman & Goldsmith) = model that states that a respondent strives to provide as
much information as possible as long as the perceived probability of both correctness and informativeness satisfy a
reasonable minimal level. According to this model, there are two theoretical knowledge states:
Satisficing knowledge (SK) = a state in which the level of knowledge is sufficient to allow oneself to provide
an answer that simultaneously satisfies both the confidence and the informativeness criteria.
Unsatisficing knowledge (UK) = a state in which the level of knowledge is insufficient to allow oneself to
provide an answer because the respondent is unable to satisfy both the confidence and the informativeness
criteria.
The original satisficing model and this dual-criterion satisficing model are basically the same when respondents have
SK.
In a unstatisficing knowledge state, it is preferable to respond ‘don’t know’, otherwise one of the two grain-selection
criteria must be violated. People would rather violate the correctness criterion than the informativeness criterion
because of social norms.
In the current research, the understanding of the grain control process in answering knowledge questions under
uncertainty is assessed.
PSY4019 Psychology of eyewitness testimonies 1
,In the first experiment, the influence of the level of relevant knowledge on providing answers is assessed.
Participants (N=24) completed the questionnaire that consisted of both 20 moderate knowledge (MK; reasonably
informative information could be given at a coarse level) and 20 low knowledge (LK; reasonably informative
information could be given at an unacceptably coarse level) items.
In the first phase, the free-grain phase, participants were instructed to answer as informative as possible and the
grain size was up to the participant. They also had to rate their confidence in the correctness of the given answer as
a percentage.
In the second phase, the fixed-grain phase, participants had to complete the same questionnaire and it was required
to answer each question at a predefined interval with (MK). Confidence in the correctness had to be rated in this
phase as well.
Results:
Accuracy for MK questions was significantly higher than for LK questions, regardless of the instructed grain
size.
Participants were overconfident in their correctness, especially for LK questions.
The confidence criterion is sometimes violated to maintain a reasonable level of informative, especially
when knowledge is low.
Thus, the dual-criterion satisficing model is supported whereas the original satisficing model is violated.
In the control experiment, the first experiment was controlled for the elicited confidence rate.
The procedure was the same as in the first experiment, however, confidence in the correctness was only elicited in
the fixed-grain phase and not in the free-grain phase.
Results:
Accuracy was the same as in the first experiment.
Thus, the results of the first experiment were not biased by the elicitation of confidence judgements in the free-grain
phase.
In the second experiment, it was tried to verify that the results in comparing LK and MK items reflect a difference in
knowledge, rather than another cause.
Participants (N=24) were presented 20 text segments in an incidental-learning phase. Afterwards, they completed
the questionnaire that consisted of 40 LK items based on the information they learned in the incidental-learning
phase in the free-grain phase followed by the fixed-grain phase. For both phases, confidence in the correctness had
to be rated.
Results:
Main effect: the accuracy of fixed-grain answers to MK items was significantly higher than the accuracy of
fixed-grain answers to LK items.
Main effect: the confidence in the correctness of fixed-grain answers to MK items was significantly higher
than the confidence in the correctness of fixed-grain answers to LK items.
Thus, performance differences and confidence are caused by different levels of knowledge instead of random noise.
In the third experiment, it was assessed if a ‘don’t know’ option is used in a free choice of grain size to resolve UK-
conflict state.
Participants (N=24) completed the same procedure as in the first experiment. However, in the free-grain phase, the
option to respond ‘don’t know’ was added. When the ‘don’t know’ option was selected, participants did not have to
rate their confidence in the correctness of their answer.
Results:
Don’t know responses were given significantly more to LK questions than to MK questions.
Thus, the dual-criterion satisficing model is supported.
Evans & Fisher (2011) – Eyewitness memory: Balancing the accuracy, precision and quantity
of information through metacognitive monitoring and control.
People volunteer to answer a question if the probability that the answer is correct is greater than a threshold value,
whereas a response is withheld if this assessed probability is less than the threshold value. This is a part of
metacognitive monitoring.
The answer ‘I don’t know’ might represent the attempt to tell nothing but the truth rather than the person
has a bad memory.
The precision of a response can be adjusted to maintain accuracy. In that case, withholding an answer is
unnecessary. This is often seen after a delay.
PSY4019 Psychology of eyewitness testimonies 2
,Input-bound measure of memory = the amount of information that is recalled.
Proportion of presented items that are later recalled.
Output-bound measure of memory = the accuracy of information that is recalled.
Proportion of provided responses that are correct.
Three factors with regard to information:
Accuracy
Precision
Quantity
Possible question formats:
Open-ended questions allow witnesses to report only accurate information at any level of precision
e.g. ‘Please tell me everything that you remember’
Cued questions provide a topic about which witnesses are asked to provide information
e.g. ‘Please describe the thief’s hair’
Forced-choice recognition questions require witnesses to respond to every question at a given level of
precision
e.g. ‘Did the thief have brown hair?’
In the current research, the impact of question type on the amount and quality of information is assessed. Also, the
common belief that memory accuracy declines over a retention interval is tested.
2 (retention interval: 10 minutes vs. 1 week) x 3 (question format: open-ended vs. cued recall vs. yes/no)
factorial design (N=126)
Participants were shown a 50-second mock crime video. Then, filler tasks were completed for 5 minutes. After this,
participants in the 10 minutes retention interval were interviewed, whereas the participants in the 1 week retention
interval came back for this interview one week later.
Participants were instructed to respond ‘I don’t know’ instead of guess if they were unable to answer. They were
either asked to provide a free narrative, were asked 15 cued questions, or were asked 15 yes/no questions.
Results:
Main effect: delay of the interview leads to significantly less information and fewer details reported for all
interview formats than when the interview is not delayed. This effect was only significant in the open-ended
condition. For the cued recall and yes/no condition, ‘I don’t know’ was answered significantly more often
after a week delay than after 10 minutes.
Marginal effect: delay of the interview leads to slightly lower accuracy in comparison to no delay.
Main effect: free narrative interview condition yields significantly higher accuracy rates than the cued
interview condition.
No effect of interview format on accuracy rate.
Limitations:
Precision doesn’t have a standardised definition.
The delay implies a week, whereas in the real world it might imply a year.
Lindsay (2007) – Autobiographical memory, eyewitness reports, and public policy.
Source-monitoring framework = model that addresses the general issue of how people differentiate between
memories from different sources with regard to autobiographical memory. It states that thoughts, images and
feelings that come to mind in response to external or internal cues, will be experienced as memories if they are
attributed to past experience.
External memory cues = environmental attributions
Internal memory cues = thoughts, images and feelings that serve as attributions
False memories arise when thoughts, images and feelings from once source are misattributed to another source,
possibly due common characteristics of the mental events.
Thus, according to the SMF, accurate discrimination between sources is needed for accurate autobiographical
memory.
PSY4019 Psychology of eyewitness testimonies 3
, McCallum, Brewer & Weber (2016) – Memorial monitoring and control: How confidence and
social and financial consequences affect eyewitnesses’ reporting of fine-grain information.
Accessible and likely-to-be correct coarse grain information is often underreported whereas likely-to-be incorrect
fine grain information is often provided since specific communication is more socially appropriate.
Memorial monitoring = ability to accurately estimate the probability that a response is correct
Memorial control = ability to decide to either report or withhold a particular response
Fine grain information is more informative than coarse grain information;
Fine grain information is less likely to be accurate than coarse grain information.
By favouring fine grain information over coarse grain information, people are favouring informativeness
over accuracy.
Social context influence social norms on appropriateness.
Social contexts involved in the memory reporting process:
Response privacy = whether the memory retriever shares their responses aloud with a wider group or
records their responses privately and anonymously.
Response audience
In the first experiment, it was examined if manipulations of response privacy and audience would induce variation in
bias towards fine grain reporting and thus underreporting coarse grain information.
2 (response privacy: public vs. private) x 2 (audience: high authority vs. low authority) factorial design
(N=85)
Participants viewed a 36-second mock crime film and answered 20 questions about this short movie. Participants
were informed either that they would be required to read their responses aloud in front of the person scoring their
items and the other participants (public) or that all answers would remain private and anonymous (private).
Participants were informed either that the police was involved in the experiment and that police interviewers would
score their responses (high authority) or that student research assistants would score their answers (low authority).
In the first phase, participants had to provide a fine grain answer and a confidence rating for every question, as a
coarse grain answer and confidence rating.
In the second phase, participants were shown their phase 1 answers without their confidence rating, and had to
select one of their two answers as their final answer to the question.
Results:
Main effect: high confidence rates predicted significantly higher accuracy, so poor monitoring resolution is
not the mechanism underlying the avoidance of reporting coarse grain information.
Main effect: coarse grain information is significantly more likely to be provided in public, whereas fine grain
information is significantly more likely to be provided in public.
No significant effect of audience.
No significant interaction between response privacy and audience.
In phase 2, fine grain information was more often selected as the final answer than coarse grain
information.
In the second experiment, it was examined if the bias to report fine grain information as found in the first
experiment, would remain existing if the willingness to risk accuracy is reduced. This willingness was reduced by
inventing a financial penalty for inaccuracy.
2 (response privacy: private vs. public) x 2 (penalty for inaccurate reporting: no penalty vs. penalty) factorial
design (N=120)
Participants viewed the same film as in the first experiment, but they only answered 10 questions about this short
movie.
Participants were informed about public response privacy as in the first experiment, whereas the private group was
instructed that they were linked to an identification number unknown by the researcher and the reimbursement
would be put in an envelope that had the same identification number.
Participants were informed either that they would receive $15 at the end of the experiment (no penalty) or that they
would start the study with a credit of $15 reimbursement but would lose $1 for every inaccurate response thus in
the end they would have a $5 - $15 payment (penalty).
The procedure was the same as in the first experiment.
PSY4019 Psychology of eyewitness testimonies 4
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper ASteijvers. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €10,49. Je zit daarna nergens aan vast.
Is Stuvia te vertrouwen?
4,6 sterren op Google & Trustpilot (+1000 reviews)
Afgelopen 30 dagen zijn er 67096 samenvattingen verkocht
Opgericht in 2010, al 14 jaar dé plek om samenvattingen te kopen