One of the primary forms of evidence collection and prosecution used by law enforcement has always been eyewitness testimony and identification of suspects. Historically, this has been a highly convincing form of evidence, with only a signed confession being more valuable in court. Until the advent of DNA testing and other forensic examinations, eyewitnesses were largely the sole proof of events occurring during a crime. However, even after the implementation of new technologies, eyewitness testimony is often used to convict and as evidence in court. They play a key role in the investigation and prosecution of crime. Despite the common misconception that eyewitness testimony is accurate and reliable, research indicates that it is often circumstantial, inconclusive, biased, or outright false, based on a range of psychological and social factors that distort the perception of events.
History and Background
Eyewitness testimony consists of the use of eyewitnesses to identify a suspect as a perpetrator of a crime. It is a “form of direct testimonial evidence used for forensic means to establish facts of an investigation or prosecution” (Albright, 2017, p. 7758). In an attempt to avoid bias and false identification, the lineup was introduced as early as 1860. Witnesses are presented with a lineup of five potential suspects from which they attempt to identify the perpetrator. This exercise also confirms the eyewitness testimony to the level where it can count as evidence in a trial. Another common use of eyewitness testimony is during the trial, witnesses are produced to testify under oath of any relevant details to the trial, such as characteristics of the suspect, known observations, or events leading up to or during the crime. Historically, eyewitness testimony was the only form of proof available as evidence. Therefore, having an eyewitness was highly valued and could easily make up for the lack of physical evidence (Nayak & Khajuria, 2019). Eyewitness testimony remains an integral part of the criminal justice system around the world.
The rise of forensic science and social sciences has shown the various dangers and negative consequences of solely relying on eyewitness testimony. Eyewitness accounts are founded on human memory and discernment, which can be easily manipulated, distorted, or distracted, potentially leading to erroneous identification or perception of events. Eyewitness testimony continues to be in use today but under different circumstances. The first changes occurred due to the 1977 U.S. Supreme Court ruling Manson v. Brathwaite establishing the modern federal framework for judicial review of eyewitness identification, where judges should evaluate the reliability of eyewitness identifications with previous rulings. In the last decade, many states have adopted stricter and more modern approaches to eyewitness testimony to deter unnecessarily suggestive identification procedures and minimize risk, finding Manson v. Brathwaite insufficient (Committee on Scientific Approaches…et al., 2018).
Variables Influencing Eyewitness Performance
All human capacity to be a witness is based on gathering, interpreting, storing, and recalling information, thus heavily dependent on the brain systems that oversee sensation, memory, and perception. Research in eyewitness science has traditionally focused on variables that impact the ability of the witness to correctly recall a memory or identify a suspect, known as eyewitness ‘performance.’ There are two categories of variables. The first is estimator variables that encompass aspects such as viewing conditions (lighting, distance, speed of subject), presence of distracting stimuli (gunshots, bright lights, glare, traffic), and the observer’s internal state, which are motivation, skill, and prejudice. The criminal justice system nor the eyewitness have direct influence over these, but they should be taken into account. Then, there are system variables, which are controlled by criminal justice actors. These range from investigation and interview tactics to how and which lineup is presented to instructions given and misinformation (Albright, 2017).
A major theme in eyewitness research and its applicability has been shifting the system variables to avoid potential influences or errors. For example, using an administrator with no knowledge of which participant is considered suspect in the lineup and standardized witness instructions for consistent response. Meanwhile, new innovations in lineups have been introduced, such as a sequential lineup where suspect faces are revealed one at a time rather than simultaneously. Both theoretically and in research, this has shown to be effective in reducing erroneous identification (Albrigh, 2017). Future research may determine how other lineup combinations or approaches to witness evidence gathering can further improve the reliability and utility of testimony. However valuable, the research does not address the underlying cause of the human factor as an observer failure with faulty perception and memory.
Psychology and Memory
There are many gaps in eyewitness accounts stemming from psychological factors—vision, which is imperfect no matter how good it is medically. There are distractions and natural noise and variations in structure from the physical environment, some distorting “optical and neuronal processes involved, some reflecting sensory content that is not relevant to the observer” (Albright, 2017). In the end, it creates multiple gaps, which are then filled by bias. Bias fills in the blanks when the perceived is unclear, using previous experience as a source. Therefore, the same system that provides certainty of perception can provide the wrong information in a way that the individual would never be able to realize, such as the effective work of our brains. Similarly, memory may be called a malleable artifact. When information is received, it is stored in memory to be used when needed to identify a suspect. Declarative memory, which is responsible for this function, consists of semantic and episodic content. Like vision, memory is influenced by uncertainty, bias, and confidence when sequenced over time (Albright, 2017).
One known phenomenon that provides an example of memory bias is the Misinformation effect. Taking a simple example, if the witness sees a green light at the intersection but is later told by another source that it was red, they are likely to testify with confidence that the light was red. The question arises as to what happened to the original memory. It is a phenomenon still being researched, but the question inherently addresses one of the fundamental concepts of memory. Researchers believe that the information received post-event weakens the past memory. Others suggest that if the misinformation is presented soon after the event, the first memory is never encoded. The brain, confused between conflicting information, chooses the second detail (Loftus, 2019). Considering the high-pressure situations that witnesses are put in as well, that is possible that testimony can be altered with misinformation, but it leads to the rise of significant concerns for validity.
Confidence in one’s memories is not an ideal predictor of accuracy, but research indicates that there is a positive relationship between confidence judgments and memory accuracy (Gustafsson et al., 2019). Two theories exist on how people can judge their own memories but have contributed to evaluating the memories of others. The theories suggest that people rely on indirect cues (heuristics) rather than accessing the memory itself when assessing its accuracy. The first theory is reality monitoring, suggesting that real and imagined memories differ in characteristics. Real memories tend to have greater contextual, sensory, and semantic details, while non-real memories are likely to refer to cognitive operations. Techniques have been developed to identify patterns in testimony in order to evaluate the veracity of memory. The second theory is cue-utilization, which argues that judgment of own memories is based on the belief of how memory works (information or theory-based) or experiences of the retrieval process (experience-based). The cues provide the strength of experience from which the strength of the memory is determined, such as a memory arising rapidly represents a strong memory representation and can be viewed as more accurate (Gustafsson et al., 2019). A cue which predicts both accuracy and confidence is response latency (Gustafsson et al., 2019). The speed with which the memory is produced, as well as the individual’s effort and confidence in the memory, are correlated with its accuracy.
Social factors must also be considered in eyewitness science. Davis and Loftus (2018) identified three characteristics of the target most commonly investigated. These are distinctiveness of appearance, available view of target features, and similarity of the suspect to the witness in race, age, and gender. Many people are very familiar-looking when not looking at the details, and with ‘change blindness’ for most typical observers, the focus on faces is superficial, typically encoding only major characteristics such as race, gender, or approximate age. The attention gap is so large that when a person is replaced during some interaction with a stranger, more than half of individuals do not notice. When an individual has exaggerated features such as unusual hair color, tattoos, scars, or disfigurement, they are easier to recognize, but at the same time, they can create unwanted attention, causing bias.
In terms of race and gender, research typically indicates that eyewitnesses are more accurate in identification targets similar to them. What is known as the own-race bias (ORB) or the cross-race effect (CRE) is essentially the impairment of identification for eyewitnesses for targets of a different race relative to one’s own race. When focusing on people of their own race, people have more correct identifications and fewer false identifications. However, when CRE comes into place, false identifications spike. The memory for the location where this member of a different race falters, and ‘change blindness’ increases. Ironically, there is also a worse self-insight regarding accuracy in recognition for cross-race individuals (Davis & Loftus, 2018). However, surprisingly, racial prejudice and bias in studies conducted over the years have little to no impact on eyewitness identification or testimony in the United States.
Consequences of Eyewitness Testimony
Eyewitness testimony can have a powerful impact on jurors during a trial. Some elements which contribute to said impact were identified by Sheahan et al. (2017). First, familiarity between the eyewitness and the juror often results in greater emphasis being put on testimony, familiarity being anything from prior contact to the association by lifestyle, gender, or race. A second factor is descriptor discrepancy; if the eyewitness describes the suspect accurately prior to the trial, jurors are more likely to believe their testimony (Sheahan et al., 2017). Eyewitness testimony is positioned by either party as the most convincing evidence, but jurors commonly err in judgment when considering the validity and integrity of such testimony. The credibility and guilt assessment of the jurors are influenced by various factors surrounding eyewitness testimony. As well-known, mistaken identification is the most prevalent factor of wrongful convictions.
Further Utilization by Law Enforcement and Court
Eyewitness identification and testimony are likely to remain within the criminal justice system as they oftentimes provide critical context to cases. However, it has been established and reiterated to both law enforcement and jurors that eyewitness accounts are unreliable so should be valued below all other types of evidence, particularly forensic and DNA evidence which is highly reliable. The American Psychological Association (2020) published guidelines that could be used in law enforcement and prosecution to improve eyewitness identification. This includes:
- interviewing witnesses as soon as possible after the crime,
- placing a suspect in a lineup should be justified in writing for that specific crime,
- lineups of several people should be used rather than a showup with a single subject,
- providing clear instructions to the witness,
- confidence statement collection from the witness,
- recording all witness interactions on video.
Most of the recommendations and policies on the matter of witness testimony are based on psychological science to ensure as little potential for error as possible. At the same time, some researchers argue that even reliable tests such as DNA can be contaminated if proper procedures are not followed. They suggest that eyewitness memory can be reliable if proper procedure is followed in testing and collecting data, such as not introducing misinformation elements. Effectively, the process has been tainted by other actors in the criminal justice system, not the eyewitnesses who provide reliable evidence on uncontaminated memory (Wixted et al., 2018). While even if being true, it is questionable how practical it would be to eliminate all influences in all cases, given how easily malleable human memory can be.
As discussed earlier, the federal Supreme Court needs to match efforts made by select state judiciaries in supporting the need for court-established practices and methods. A national policy is needed for law enforcement guidance in collecting eyewitness identifications and their application in court, having been evaluated through known psychology theories (Azar, 2011).
Ironically, eyewitness testimony directly impacts Christianity because, in theory, the Gospels and all events in them are eyewitness testimony. Therefore, given the lack of reliability for eyewitnesses, should the Gospels be trusted? It is an issue that the apostles faced for many years as the Bible came into place. However, there are some standards, such as whether the individual is present during the event, whether the evidence is corroborated by other sources or eyewitnesses, and whether the evidence is consistent and accurate. For the most part, these fit the apostles and their writings, and although there are some differences in the version of events, they are minor and work to support one another.
The Bible supports eyewitness testimony, but only if corroborated. It is written, “Whoever is deserving of death shall be put to death on the testimony of two or three witnesses; he shall not be put to death on the testimony of one witness” (New International Version, 2011, Deut. 17:6-7). At the same time, Deuteronomy 13:14 talks about investigating carefully and diligently when searching for evidence of a crime, and Exodus 20:16 warns about bearing false witness. Finally, John 19:35 states, “And he who has seen has testified, and his testimony is true; and he knows that he is telling the truth, so that you may believe.” Similar to the evidence and discussion in this paper, Christianity understands the weight and consequence of eyewitness testimony regarding judgment and punishment. Therefore, Christians are called upon to never assume or be aware of the deceit, but search for corroborating evidence and accounts to ensure justice is truly served, and an innocent person does not suffer.
Eyewitness identification and testimony have historically been a consistent method of investigation and viewed as highly reliable evidence. However, the advent of DNA technologies, as well as recent research into memory and psychology, demonstrate significant flaws in eyewitness accounts. From a criminal justice perspective, this is consequential as it leads to innocent people receiving punishment, sometimes even the death penalty, while also undermining the public trust in the system. Eyewitnesses can be potentially valuable, particularly immediately after a crime is committed to guide investigators, but as a piece of evidence it should be disallowed in court. There are too many factors to consider such as personal cognitive capacity and memory, as well as the traditional ‘hearsay’ and motives. Although already guidelines and standards to eyewitness reporting,
Albright, T. D. (2017). Why eyewitnesses fail. Proceedings of the National Academy of Sciences, 114(30), 7758–7764.
American Psychological Association. (2020). Improving eyewitness identification key to protecting innocent people.
Azar, B. (2011). The limits of eyewitness testimony. Monitor on Psychology, 12.
Committee on Scientific Approaches to Understanding and Maximizing the Validity and Reliability of Eyewitness Identification in Law Enforcement and the Courts, Committee on Science, Technology, and Law; Policy and Global Affairs, Committee on Law and Justice; Division of Behavioral and Social Sciences and Education, National Research Council. (2018). Identifying the culprit: Assessing eyewitness identification. The National Academies Press.
Davis, D., & Loftus, E. F. (2018). Eyewitness science in the 21st century. Stevens’ handbook of experimental psychology and cognitive neuroscience, 1–38.
Gustafsson, P. U., Lindholm, T., & Jönsson, F. U. (2019). Predicting accuracy in eyewitness testimonies with memory retrieval effort and confidence. Frontiers in Psychology, 10(703).
Loftus, E. F. (2019). Eyewitness testimony. Applied Cognitive Psychology, 33(4).
Nayak, B. P., & Khajuria, H. (2019). Eyewitness testimony: Probative value in criminal justice system. Egyptian Journal of Forensic Sciences, 9(1).
New International Version Bible. (2011). Biblica, Inc.
Sheahan, C. L., Pozzulo, J. D., Reed, J. E., & Pica, E. (2017). The role of familiarity with the defendant, type of descriptor discrepancy, and eyewitness age on mock jurors’ perceptions of eyewitness testimony. Journal of Police and Criminal Psychology, 33(1), 35–44.
Wixted, J. T., Mickes, L., & Fisher, R. P. (2018). Rethinking the reliability of eyewitness memory. Perspectives on Psychological Science, 13(3), 324–335.