Mistaken eyewitness identifications have contributed to approximately 75% of false convictions. Meaning that more than 100 people each year could be falsely convicted of violent or sexual crimes because of these false accusations from witnesses (Garrett, 2011). Eyewitness misidentification is believed to be the single greatest cause of wrongful convictions nationwide which are later overturned through DNA testing.
Research studies since the 1960s (Brandon & Davies, 1973; Huff; Rattner & Sagarin, 1986) have demonstrated that there are significant reasons to question the accuracy of eyewitness testimony used in court trials. Eyewitnesses are often very confident in the accuracy of their memories when identifying a suspect, but in reality, the malleability of human memory and individual perception makes eyewitness testimony a very unreliable source of evidence (Garrett, 2011). Forensic science can disprove the myth that memory provides an accurate recording of experience, like a video camera and instead support psychologists in their claim that people’s memories and visual perception can even be manipulated and biased. Due to this, many countries are now attempting to make changes in how eyewitness testimony is presented in court. But, despite a concrete and growing amount of evidence demonstrating the faults of traditional eyewitness identification procedures and the availability of basic strategies to reform them (Innocence Project, 2009). Research still shows that people naturally rely on witnesses’ statements about their confidence, view, and the degree of attention they paid while witnessing to make judgments about whether to believe the witness (Bradfield & Wells, 2000). Therefore, the purpose of this report is to outline which problems influence the accuracy of eyewitness identification and prove that other procedural measures must be undertaken to minimize the likelihood of an inaccurate identification.
The most common line-up procedure used in the criminal justice system is having all of the lineup members are displayed to the eyewitness at once in a simultaneous lineup. However, during a simultaneous lineup witnesses use a ‘relative judgment’ process, meaning that they compare lineup photographs or members to each other, rather than to their memory of the offender. This is a problem when the perpetrator is not present in the lineup because often the witness will choose the lineup member who most closely resembles the perpetrator (Wells & Seelau, 1995). Research instead suggests law enforcement should implement a sequential line-up procedure (Carlson, Gronlund & Clark, 2008), where the witness is shown only one person at a time and are required make a ‘yes’, ‘no’, or ‘not sure’ response to each one before moving on to the next. Lindsay, Lea, Nosworth, Fulford, Hector, LeVan & Seabrook (1991) conducted five staged-crime experiments to examine the effect of lineup biases and sequential presentation on eyewitness recognition accuracy. They discovered across their five experiments biased lineup procedures consistently produced higher rates of false identification from simultaneous presentations (51.0%) than from sequential presentations (12.9%). Correct rejections of the biased, criminal-absent lineups were less likely in simultaneous presentations (29.0%) than in sequential presentations (77.1%). Even when the lineups were not biased, false identifications occurred more often with simultaneous (23.3%) than sequential (4.2%) presentations.
These results support the idea that the sequential procedure is more conservative than the simultaneous procedure. The reason for this is because the psychological experience for the eyewitness is dramatically different using the sequential procedure than it is using the simultaneous procedure. Using the sequential procedure, the eyewitness cannot simply compare one photo to another and decide who looks the most like the offender relative to the others. (Klobuchar & Caligiuri, 2005). Although the eyewitness can mentally compare the current photo to those presented previously, the eyewitness cannot be sure what the next photo will look like; maybe the next one will look even more like the offender. The sequential procedure is conducted such that the witness does not know when the last photo is being shown. As a result, the sequential procedure appears to lead eyewitnesses to set higher criteria for making a positive identification because they cannot be sure that they have seen all of the lineup members.
Another recommendation for improving lineup procedures to prevent false identifications is to limit the number of show-ups, in which the police show the eyewitness one person. Show-ups tend to occur when a suspect who fits the description is found close to a crime scene shortly after the crime. However, the suspect is most frequently in handcuffs and may be showing other signs of distress for example sweating. These circumstances create a highly suggestive situation in which the risk of a misidentification dramatically increases. Yarmey, Yarmey & Yarmey (1996) tested one-person and six-person photographic lineup identifications in field situations either immediately, or 30 minutes, or 2 hours, or 24 hours after a 15-second ordinary encounter with a target. They discovered that after 2-h, the mean proportion of false identifications of an innocent suspect was four times likely with show-ups (M = .58) than photo line-ups (M = .14). Similarly, false identifications, again at a ratio of 4:1, occurred significantly more frequently in one-person line-ups (M = .47) than in six-person line-ups (M = .14), 24-h after the original encounter. This supports the claim that show-ups produce higher rates of mistaken identification than do simultaneous line-ups or sequential line-ups. The reason for the higher rate of mistaken identifications when using show-ups is mainly because show-ups do not include fillers that protect an innocent suspect (Gonzalez, Ellsworth & Pembroke, 1993).
On the other hand, line-ups and photo arrays of course, are far less suggestive; if conducted properly, the witness will not know which person the officer believes to be the true perpetrator and, therefore, will not be influenced in the identification process. If the witness makes a false identification, he will identify a subject that the police have purposely inserted as filler (Gronlund, Carlson, Neuschatz, Goodsell, Wetmore, Wooten & Graham, 2012). Consequently, the witness’s misidentification will be known to the police, who can then continue with their investigation and their search for the true perpetrator. However, there are problems within traditional line-up procedures that need to be reformed as well in order to reduce misidentification.
Another cause of inaccurate identification within police line-ups are ones that do not include the perpetrator of the crime. When the actual perpetrator is not included in the line-up, the police suspect faces a significantly heightened risk of being incorrectly identified as the culprit. Wells’ (1984) meta-analysis of issues and data on eyewitness identification discovered that witnesses who might correctly identify a culprit from a culprit-present lineup will misidentify an innocent suspect from a culprit-absent lineup, leading to a significant increase of incorrect identifications from 20% to 70%. This suggests that eyewitnesses are most likely a relying on a ‘relative judgement’ process again (Sobel, 1972; Clifford & Bull, 1978). Whereby, they tend to select the person who looks ‘most like’ the offender. When the actual perpetrator is not present in the line-up, the police suspect is often the person who best fits the description, hence his or her selection for the line-up (Doob & Kirshenbaum, 1973).
Thus, since eyewitnesses approach line-ups with the goal of finding the offender. They should be cautioned that the offender might not be in the line-up because they need to understand that they are not ‘failing’ if they do not choose someone. Studies have shown that instructing a witness that the perpetrator ‘may or may not be present’ in the line-up can dramatically reduce the likelihood that a witness will identify an innocent person. Whereby, Malpass and Devine (1981) found that failure to warn witnesses that the culprit may or may not be in the line-up resulted in 78% of witnesses making identifications from a target absent line-up, whilst maintaining a high level of hits in the target present line-up. With the warning that the perpetrator may not be in the target absent line-up the false identification rate fell to 33%. Therefore, providing a cautionary instruction legitimizes a ‘no choice’ selection for the eyewitness who might otherwise select the individual who most closely resembles the perpetrator. If the perpetrator is absent, because the suspect in the lineup is actually an innocent person, the use of a cautionary instruction thus lessens the chance of a mistaken eyewitness identification.
In standard line-ups as well, a set of ‘fillers’ will accompany the suspect in the line-up whereby the line-up administrator may choose to compose a live or photo line-up where non-suspect ‘fillers’ do not match the witness’s description of the perpetrator or do not resemble the suspect. This can cause the suspect to stand out to a witness because of the composition of the line-up. A meta-analysis by Clark, Howell & Davey (2008), of a staged-crime experiment suggests that approximately 50% of witnesses presented with a perpetrator-absent line-up select an innocent suspect. Furthermore, Valentine, Pickering & Darling (2003) analyzed data from the Metropolitan Police and in 640 attempts by eyewitnesses, to identify the alleged culprit in 314 line-ups, approximately 33% of those identifications included a known-innocent filler. This shows that placing an innocent suspect who fits the description of the offender in a line-up in which the fillers do not fit the description, results in a high rate of mistaken identifications of that person, even when absolute similarity between the innocent person and the offender is only moderate (Carlson, Gronlund, & Clark, 2008).
Research instead suggests that suspect photographs should be selected that do not bring unreasonable attention to him/her. Non-suspect photographs and/or live line-up members should be selected so that the suspect does not stand out from among the other fillers. Lindsay & Wells (1980) found in a stage crime paradigm that in the culprit-absent line-ups, there was a significant difference in the proportion of choices for the innocent suspect between low (.70) and high (.31) line-up similarity. Therefore, law enforcement should select fillers using a blended approach that considers the fillers’ resemblance to the description provided by the eyewitness and their resemblance to the police suspect, to improve the accuracy of eyewitnesses.
Another factor that can cause misidentification is investigator bias. In a standard line-up, the line-up administrator typically knows who the suspect is, and research has shown that administrators often provide biased unintentional cues to the eyewitness about which person to pick from the line-up (Clark, 2005; Steblay, 1997). Devenport & Fisher (1984) examined the effect of biased instructions alongside the presence of authority where in different conditions the line-up administrator was dressed casually or as a police officer. They discovered, in culprit-absent line-ups, a difference in the proportion of choices for the foils between unbiased (.471) and biased instructions (.706). This proportion increased further when the investigator gave biased instructions as a police officer; .800. This supports the idea that the investigators knowledge of the line-up can influence an eyewitness’ decision in choosing suspect. Hence, since 92.1% of live line-ups are conducted by an administrator who knows the identity of the suspect (Police Executive Research Forum, 2013), the ideal solution is to prevent any opportunity for administrators bias that could lead to intentional or unintentional cues.
One was to do this, is implanting a double-blind procedure, where neither the administrator nor the eyewitness knows who the suspect is. This prevents the administrator of the line-up from providing inadvertent or intentional verbal or nonverbal cues to influence the eyewitness to pick the suspect. In Greathouse & Kovera’s (2009) study, witnesses and line-up administrators completed an identification task in which administrator knowledge, line-up presentation, instruction bias, and target presence were manipulated. They discovered that single-blind investigators were more likely than double blind witnesses to examine the lineup carefully (F(1,144) = 6.88), to tell witnesses that they know who the suspect is (F(1,144) = 7.41), to tell witnesses to take another look at the lineup if they did not make an identification (F(1,144) = 5.54), and to remove a picture slowly if witnesses said no to a particular photograph, F(1,144) = 3.91). Therefore, research shows that double-blind lineup procedures are needed not only to prevent the investigator from unintentionally influencing which person the eyewitness picks, but also are needed to prevent the investigator from influencing the certainty of the eyewitness.
Another area in which investigator bias can appear is post-identification feedback, this might be partially to blame for wrongful convictions because it inflates eyewitnesses’ reports of their confidence and other testimony-relevant eyewitness reports (Wells & Bradfield, 1998). Steblay, Wells & Douglass (2014) conducted a meta-analysis of studies involving over 7000 participant-witnesses. They discovered that the percentage of mistaken witnesses who reported confidence levels of 80% increased from 6% to 29% when they were given feedback following their mistaken identifications. This supports the claim that suggesting to eyewitnesses who mistakenly identify someone from a lineup that they identified the right person leads them to recall having been more certain, having a better view during witnessing, and having paid closer attention during witnessing. Furthermore, confirming feedback can even distort evaluators’ abilities to discriminate between accurate and mistaken eyewitness testimony.
Therefore, at the time an eyewitness makes an identification, a statement should be obtained from the eyewitness in a double-blind procedure, indicating how confident he or she is that the person identified is the offender. This will allow investigators to assess the confidence of the eyewitness before that person can be influenced by other events, such as learning the status of the identified person. A recent survey of law enforcement agencies by the Police Executive Research Forum (2013) found that 76.2 percent document statements of certainty related to a positive identification. These confidence statements can be strengthened through preparing a complete and accurate record, audio or visual, of the outcome of the identification procedure. This will improve the strength and credibility of the identification or nonidentification results obtained from the witness (Du Toit, De Jager, Paizes, Skeen & Van der Merwe, 1987). Hence, if the eyewitness was in fact uncertain at the time of the identification, then this record should be made aware to all agents involved in the criminal justice system.
So, in conclusion, mistaken identifications are the primary cause of convicting the innocent and if implemented, these recommendations for line-ups would advance justice in several ways. First, these recommendations would benefit innocent suspects because they would be less likely to be misidentified. Second, implementation of these procedures would help keep the focus of investigations on guilty persons. Third, implementation of these procedures would be of assistance to those involved in evaluating the identification testimony such as prosecutors, judges, and jurors. Finally, the use of these procedures would lessen the need for expert testimony because expert testimony commonly focuses on the inadequacies of the procedures used in line-ups. Therefore, reforms to how eyewitness identification evidence is collected, preserved, and used must take center stage within the UK government’s institute of justice, to improve the criminal justice system and prevent the conviction of the innocent as well as facilitating conviction of the guilty.