ewt reading Flashcards
(34 cards)
Simons and Chabris (2011)
found that 37% of Americans believe the testimony of a single confident eyewitness is sufficient to convict a criminal defendant. However, the increased use of DNA testing in recent years has suggested that there are significant dangers associated with relying on eyewitness identification because more than 200 individuals convicted on the basis of mistaken eyewitness identification have been proved innocent by DNA tests.
Charles Chatman
He was 20 years old when a young woman who had been raped picked him out from a lineup. As a result of her eyewitness testimony, Chatman was sentenced to 99 years in prison in Dallas County, Texas. DNA testing led to Chatman being released after 26 years in prison. Chatman claimed that race was a factor: “I was convicted because a black man committed a crime against a white woman.”
Smalarz and Wells (2012)
estimated that in only approximately 5% of cases is DNA evidence potentially available that might show eyewitnesses have identified the wrong person.
DNA tests are not infallible.
They can indicate that a given individual was present at the scene of the crime but not necessarily that he/she actually committed the crime.
Factors that influence the accuracy of eyewitness testimony
Garrett 2011 - 161 cases
where DNA evidence indicated that mistaken eyewitness identification had led to the conviction and imprisonment of innocent individuals. = eyewitnesses have fallible memories and their confidence in identifications should be disregarded eg case of ronald cotton - Jennifer Thompsonbecame progressively more confident she had identified the culprit when she received positive feedback from the police following her initial identification.
Steblay, Wells, and Douglass (2014)
carried out a meta-analytic review based on approximately 7,000 participants of the effects of such feedback. There was a strong tendency for participants to remember mistakenly they had had been very confident of the accuracy of their identification prior to receiving the positive feedback. This is the post-identification feedback effect.
we can trust eyewitnesses’ confidence in their identifications provided we consider only their initial level of confidence.
Wixted, Mickes, Dunn, Clark, & Wells, 2016 -
Only approximately 20% of eyewitness identifications of culprits were correct when their confidence was low. In contrast, the corresponding figure was 80% when their confidence was high.
Judges except ewt to be accurate -
Wise and Safer (2004) found American judges substantially underestimated the importance of factors causing eyewitness testimony to be inaccurate. As a result, 77% of judges were willing to accept that a defendant should be convicted of a crime based solely on eyewitness testimony. Wise and Safer (2010) found American judges’ knowledge of factors influencing the accuracy of eyewitness testimony was comparable to that of undergraduate students. The findings strongly suggested that the number of wrongful convictions based on eyewitness testimony would be reduced if judges had greater relevant knowledge.
Change blindness -
Change blindness is a related phenomenon that also depends on attentional limitations. It involves a failure to detect changes in an object - its very common. hange blindness can cause problems when eyewitnesses try to remember an event. While it would appear that change blindness is an unfortunate defect, that is not necessarily the case. Fischer and Whitney (2014) argued that our visual world is typically relatively stable over short time periods. As a result, it is worthwhile for perceptual accuracy to be sacrificed occasionally (as in change blindness) so we have continuous, stable perception of our visual environment.
Change blindness blindness -
Suppose we presented participants with videos previously used to show change blindness and asked them whether they would personally have detected the changes. Precisely this has been done in several studies (e.g., Jaeger, Levin, & Porter, 2017; Levin, Drivdahl, Momen, & Beck, 2002). It has been found consistently that participants massively overestimated their ability to detect changes. This phenomenon is known as change blindness blindness.
Levin et al. (2002). Participants saw videos of two people chatting in a restaurant. In one video, the plates on their table changed from red to white, and in another a scarf worn by one of them disappeared. A third video showed a man sitting in his office and then walking into the hall to answer the telephone. When the view switches from the office to the hall, the first person has been replaced by another man wearing different clothes.
The above videos had previously been used by the researchers, who found no participants detected any of the changes. Levin et al. (2002) asked their participants to indicate whether they thought they would have noticed the changes if they had not been forewarned. The percentages claiming they would have noticed the changes were as follows: 78% for the disappearing scarf; 59% for the changed man; and 46% for the change in color of the plates.
When we look at the environment, we obtain only limited information from peripheral vision. However, we often use top-down processes (e.g., expectations) to fill in the gaps in the information available to us. As a consequence, “We see far less than we think we see” (Cohen, Dennett, & Kanwisher, 2016, p. 324)
Expectations - Hastorf and Cantril’s (1954) findings show confirmation bias—
event memory is influenced and systematically distorted by the observer’s expectations. More evidence of confirmation bias was reported by Lindholm and Christanson (1998). Swedish and immigrant students watched a videotaped simulated robbery where the perpetrator seriously wounded a cashier with a knife. The perpetrator was either Swedish (blond hair and light skin) or an immigrant (black hair and brown skin). The key finding was that both immigrant and Swedish eyewitnesses were twice as likely to identify as the culprit an innocent immigrant as an innocent Swede from color photographs. Immigrants are over-represented in Swedish crime statistics, and this influenced participants’ expectations about the perpetrator’s likely ethnicity. Barlett 1932 explained why this occurs - we possess numerous schemas or packets of knowledge stored in long-term memory. These schemas lead us to form certain expectations. A major prediction from Bartlett’s theory is that eyewitness memory should often be distorted to conform to the relevant schema.
Misinformation effect -
they often fail to pay sufficient attention to the crime and the criminal(s). After all, the crime they observe typically occurs suddenly and unpredictably. However, Loftus and Palmer (1974) argued that what matters is not only what happens at the time of the crime. According to them, eyewitness memories can surprisingly easily be distorted by what happens after observing the crime.
Loftus and Palmer (1974) asked their participants to watch a film of a car accident. Afterwards, they described what had happened and then answered specific questions. Some participants were asked, “About how fast were the cars going when they hit each other?” Others were asked the same question but with the word hit replaced by collided, bumped, contacted, or smashed into.
What did Loftus and Palmer (1974) find? Speed estimates were highest (40.8 mph) when the word smashed was used, lower with collided (39.3), and lower still with bumped (38.1 mph), hit(34 mph), and contacted (31.8 mph). One week later, all participants were asked, “Did you see any broken glass?” There was no broken glass, but 32% of those previously asked about speed using the verb smashed said they had seen broken glass. In contrast, only 14% of the participants asked using speed using the verb hit claimed to have seen broken glass.
Thus, our memory for events is so fragile it can be systematically distorted by changing one word in one question! This exemplifies the misinformation effect—misleading information presented after an event causes distortions in memory for that event. The findings discussed so far demonstrate retroactive interference (disruption of memory by the learning of other material during the retention interval between original learning and the memory test
- Eyewitness memory can also be distorted by proactive interference (learning occurring prior to observing the critical event
There is compelling evidence that the accuracy of long-term memory generally depends on various cognitive factors (e.g., intelligence; working memory capacity: Unsworth, 2019)
Theoretical explanations -
One important factor is source misattribution (Johnson, Hashtroudi, & Lindsay, 1993). In essence, information about an event is remembered correctly but the source or context of that information is misremembered. Unsurprisingly, source misattribution is most likely to occur when the memories from one source closely resemble those from a second source. For example, Lindsay et al. (2004) presented participants with a narrative and a video. Source misattribution (intruding information from the narrative into their memory of the video) was much more common when the two events were similar.
A key theoretical issue is whether misinformation causes permanent alteration of memory traces from a witnessed event. Oeberst and Blank (2012) argued that misinformation typically does not permanently change memory traces. Instead, the misinformation effect occurs because eyewitnesses are instructed to recall the single correct account of an event.
Interventions -
Gabbert, Hope, Fisher, and Jamieson (2012) argued that an early opportunity to recall the details of a witnessed crime would protect the relevant memory traces from distortion by misleading information. Eyewitnesses were shown the video of a bank robbery. After that, some of them then recalled all the details of the crime they could remember. One week later, all participants received misinformation followed by event recall. Those who had provided immediate recall were more resistant to this misinformation.
Anxiety and violence
Weapon focus -
Weapon focus - Much of the research on anxiety and eyewitness memory has investigated the weapon focus effect—the presence of a weapon causes eyewitnesses to attend to the criminal’s weapon which in turn causes them to have reduced memory for details of the assailant and the environmental context. For example, Biggs, Brockmole, and Witt (2013) found that observers generally fixated weapons more than neutral objects. As a consequence, they fixated faces less often in the weapon condition. Pickel (2009) pointed out that people often attend to stimuli that are unexpected in a situation, which impairs their memory for other stimuli. This led Pickel to argue that the weapon focus effect will be greater when the presence of a weapon is very unexpected. As predicted, there was a stronger weapon focus effect when a criminal carrying a folding knife was female, because it is more unexpected to see a woman with a knife. Also as predicted, the weapon focus effect was greater when a criminal with a knitting needle was male rather than female.
Stress impairs memory - stress causes a narrowing of attention in which peripheral details receive little attention and are poorly remembered.
Yegiyan and Lang (2010) reported additional support in a study where they presented people with distressing pictures. As picture stressfulness increased, recognition memory for peripheral details decreased progressively whereas memory for central details was enhanced.
It is important to emphasize that the effects of stress on memory are relatively complex. For example, Quaedflieg and Schwabe (2018) discussed research showing that stress generally increases rigidity in learning and memory and reduces the involvement of episodic memory (see Chapter 6). Such rigidity reduces the extent to which contextual details are incorporated within the memory trace (as predicted by Easterbrook’s hypothesis). However, stress also reduces the ability to modify existing memories in light of new information. As a consequence, stress can reduce the misinformation effect (discussed earlier) by “protecting” memories from updating by post-event misinformation (Schmidt, Rosga, Schatto, Breidenstein, & Schwabe, 2014).
Age
- the eyewitness testimony of older adults is generally less accurate than that of younger ones. Fraundorf, Hourihan, Peters, and Benjamin (2019) reviewed 232 studies on recognition memory. Overall, the memory performance of older adults was inferior to that of younger ones. In addition, older adults were more likely to judge old items as new.
- Healey and Kahane (2016) proposed a four-component model to explain age-related differences in long-term memory. First, older adults have reduced ability to sustain attention at the time of learning (or witnessing of an event). Second, they are less able to retrieve relevant contextual information to facilitate recall. Third, they find it harder to monitor their retrievals and reject incorrect items. Fourth, the retrieval process in older adults is less precise and more likely to produce “noise.”
- Older adults exhibit greater misinformation effects than younger ones. Jacoby, Bishara, Hessels, and Toth (2005) found the presentation of misleading information caused older adults to have a 43% chance of recalling false memories compared to only 4% for younger ones. Subsequent research indicated that this large difference occurred in part because older adults are less likely to monitor their own recall to reduce errors (Morcon, 2016).
Wright and Stroud (2002) found that adults identifying culprits after viewing crime videos showed own-age bias—younger and older adults tend to show better facial recognition for individuals close to themselves in age. Martschuk and Sporer (2018) reviewed the evidence. Younger adults had generally better facial recognition memory than older ones. However, the difference was significantly less with old faces than young ones. It might be due to expertise—most of us have greater exposure to (and familiarity with) faces of individuals of our own age. Wiese, Wolff, Steffens, and Schweinberger (2013) reported supporting evidence. Young geriatric nurses had no own-age bias because they recognized old faces much better than did young controls because of their extensive experience with older people.
Remembering faces
The culprit’s face is often easily the most important information eyewitnesses may or may not remember accurately.
- Young and Burton (2018) discussed whether or not we are face experts. In essence, they argued we are experts when recognizing familiar faces. However, we find it surprisingly hard to recognize unfamiliar faces. This is a finding of direct relevance when trying to understand the limitations of eyewitnesses’ memory for the faces of culprits. However, it should be noted that eyewitnesses are often acquainted with the culprit with certain crimes (e.g., assault; rape).
We are often surprisingly poor at recognizing unfamiliar faces even when we do not have to rely on our fallible memory. Bruce et al. (1999) focused on people’s ability to identify someone on the basis of closed-circuit television (CCTV) cameras. Participants were presented with a target face taken from a CCTV video, together with an array of 10 high-quality photographs (see Figure 12.7). Their task was to select the matching face or to indicate that the target face was not present in the array. Bruce et al. (1999) found performance was disappointingly poor. When the target face was present in the array, it was selected only 65% of the time. When it was NOT present, 35% of participants nevertheless claimed one face in the array matched the target face. Allowing participants to watch a 50-second video segment of the target person as well as a photograph of their face failed to improve identification performance.
Patterson and Baddeley (1977) discovered several factors that influence our ability to recognize unfamiliar faces.
Participants were presented with photographs of individuals photographed undisguised or wearing a beard, wig, spectacles, or any combination thereof. The photographs were taken full face or in profile. The participants were familiarized with one photograph of each person in any one combination of disguised features. This was repeatedly presented until it was consistently recognized and the person’s name given correctly. Their participants were then presented with photographs consisting of the target individuals in all possible combinations of disguise, either in full frontal view or in profile, together with a number of similarly disguised but unfamiliar people. Their task was to detect and name the target individuals.
What did Patterson and Baddeley (1977) find? There was a dramatic effect of disguise. Every time an item of disguise was added or removed the probability of correct recognition decreased. Performance ranged from extremely good when the face was presented in its originally learned form to virtually guesswork when the maximum number of disguised features was changed. These findings suggest that criminals are well-advised to wear masks or other forms of disguise even if this makes them look very conspicuous!
Unconscious transference
eyewitnesses are sometimes better at remembering faces than at remembering the precise circumstances in which they previously saw a face. This can have serious consequences. For example, it can cause eyewitnesses to recognize correctly a face as having been seen before but to judge incorrectly that that person was guilty of committing a crime. This is unconscious transference.
Verbal overshadowing
- the verbal overshadowing effect—“the finding that describing a previously seen face can impair its subsequent recognition”
Schooler and Engstler-Schooler (1990) provided the first demonstration of the verbal overshadowing effect. Eyewitnesses watched a film of a crime. After that, some eyewitnesses provided a detailed verbal report of the criminal’s face, whereas others did an unrelated task. Those who had provided the detailed verbal report performed worse than the other eyewitnesses on a subsequent recognition-memory test. Subsequent research has produced somewhat mixed findings. However, Alogna et al. (2014) successfully replicated Schooler and Engstler-Schooler’s (1990) findings.
Other-race effect
- the other-race effect (sometimes known as the cross-race effect)—same-race faces are typically recognized more accurately than other-race faces (see Young, Hugenberg, Bernstein, & Sacco, 2012, for a review). This effect depends on various factors. One such factor is expertise. Unsurprisingly, eyewitnesses having the most experience with members of another race have a smaller other-race effect than those with less experience (Hugenberg, Young, Bernstein, & Sacco, 2010).
- it was assumed the other-race effect occurs because we find it hard to remember the faces of individuals of a different race. Megreya, White, and Burton (2011) showed this assumption is only partially correct because it also depends importantly on perceptual processes. British and Egyptian participants saw a target face and an array of 10 faces (see Figure 12.8) and decided whether the target face was in the array. If it was present, they identified which face it was. Of importance, this task imposed minimal demands on memory because all the photographs remained visible.
Techniques to increase memory accuracy
Increase face recognition - their recognition-
memory performance should be enhanced if information from multiple photographs of the same unfamiliar face were combined to create an average. Jones, Dwyer, and Lewis (2017) tested this prediction. Observers viewed a single front-view photograph of an individual (the target), seven photographs of that individual at different orientations, or seven computer-generated synthesized images of that individual at different orientations (see Figure 12.9). After that, the observers selected the target face from an array of five faces.
The way lineups are presented
Techniques to increase memory accuracy - Cognitive interview -
Over the past 30 years or so, the police have increasingly made use of various versions of the cognitive interview, which was originally devised by Geiselman, Fisher, MacKinnon, and Holland (1985). This approach is based on four general retrieval rules:
1Mental reinstatement of the environment and any personal contact experienced during the crime.
2Encouraging the reporting of every detail regardless of how peripheral it might seem to the main incident or crime.
3Describing the incident in several different orders.
4Reporting the incident from different viewpoints including those of other participants or witnesses.
What are the main limitations with the cognitive interview?
1The small increase of incorrect information recalled by eyewitnesses can lead detectives to misinterpret the evidence.
2It does not reduce the negative effects of misinformation (Memon, Zaragoza, Clifford, & Kidd, 2010).
3Most versions of the cognitive interview do not require eyewitnesses to recall events with their eyes closed. However, Vredeveldt, Hitch, and Baddeley (2011) found this enhanced recall, and so it should be incorporated within cognitive interviews. Eye closure is beneficial because it reduces eyewitnesses’ cognitive load and minimizes distraction.
4The cognitive interview is less effective when the event was stressful. It is also less effective when there is a long delay between the event and the interview.
