Issues of Bias in Forensic Casework: Some Causes, Effects, and Potential Solutions - Essay Example

Issues of Bias in Forensic Casework: Some Causes, Effects, and Potential Solutions Arguably, the most important goal of using forensic experts in criminal investigations is to have people with the right knowledge, skills, and perspective to assess forensic evidence, so that they can present valid, reliable, and persuasive interpretations of the latter, but in reality, these people cannot completely prevent biases from affecting their analyses. To better understand diverse studies on this topic, in a previous assignment that examined the research design, data analysis, and findings of three works from Bergman and Fox (2009), Cohen and Smith (2010), and Gianelli (2010), findings showed that quantitative research can provide valid answers to causality-based research questions. Nonetheless, qualitative research improves the richness of data through understanding the how’s and why’s of research outcomes. This essay focuses on forensic science biases, as it explores several studies on bias in forensic science and aims to understand some of its causes, consequences, and resolutions. It combines qualitative and quantitative research in answering these research questions. Cognitive bias is present in forensic science, and it must be recognized and accepted in the field, in order for it to be properly resolved. Cognitive bias pertains to the psychological tendency of supporting one opinion over another because of access to extraneous information; in other words, it is bias that comes from “knowing something” (Page, Taylor, & Blenkin, 2012, p.108). In “Cognitive Bias in Forensic Science,” Gianelli (2010) studied the impact of cognitive bias on forensic analysis through reviewing quantitative studies. After examining existing studies on cognitive bias issues, he was alarmed that experimental studies indicated that numerous forensic scientists portray widespread cognitive bias. Apparently, their expertise (when present) is not always enough to shield them from cognitive bias. He stressed the importance of conducting additional studies in understanding the causes, nature, and directions of cognitive bias. Furthermore, cognitive bias cannot be entirely avoided because forensic experts are human beings who can be psychologically and cognitive swayed by their emotions and other information or conditions. Page, Taylor, and Blenkin (2012) confirmed the existence of cognitive bias in forensic science and importance of its acknowledgment in their article, “Context Effects and Observer Bias-Implications for Forensic Odontology.” Using diverse qualitative and quantitative studies, they underscored that the “ignorance of contextual effects is a recognized psychological phenomenon in itself” and is called “bias blind spot” (p.108). They stressed that contextual bias happens, especially for odontologists who deal directly with victims and their families (Page et al., 2012, p.109). Acceptance of bias makes experts more aware of limiting their overconfidence in their analysis and decreasing the effect of bias on their interpretations by controlling potential sources of bias and how they can affect cognitive abilities (Page et al., 2012, p.108). Contextual information can affect cognitive bias among forensic experts. Dror (2012) pursued a quantitative research design in “Cognitive Forensics and Experimental Research about Bias in Forensic Casework” to ask the same question about cognitive bias that Gianelli (2010) did. Dror (2012) compared the decisions from seventeen DNA analysts who studied a DNA mixture without biased contextual data with two forensic examiners who evaluated the same DNA mixture within biased contextual data of a real criminal case. With merely one of the examiners in the control “non-biased” group arriving at the conclusion of “cannot be excluded,” while two examiners in the “biased” condition said “cannot be excluded,” findings showed that extraneous information can result to cognitive bias. Contextual information can impact cognitive bias, which affected DNA mixture analysis. Moreover, forensic scientists may be susceptible to contextual information when they are constant access to different groups with diverse, sometimes opposing, interests and opinions. Dror and Rosenthal (2008) assessed the reliability and biases of forensic experts in “Meta-analytically Quantifying the Reliability and Biasability of Forensic Experts.” Using two types of studies, they learned that forensic experts are not flawless in their analysis: The first two studies to examine these questions established that experts are far from being perfect. These studies demonstrated circumstances in which experts were both relatively unreliable and biasable, and in the analyses reported here we quantify these effects statistically and subject them to meta-analytic procedures. The data are based on forensic decision making made by latent fingerprint experts, but because this forensic domain is the most widely used and well established, we can be confident that the problems exposed within this domain are also prevalent in other forensic domains. (Dror & Rosenthal, 2008, p.903). Though the study cannot generalize across forensic domains, it shows that even experts make mistakes. These mistakes must be acknowledged, so that they can be better managed in the future. This paper ascertained that when forensic scientists interact with different people, the more that the latter can affect their analysis and interpretation. Different exposures, for instance, can result to different interpretations. This thread of research deserves further scrutiny in future qualitative and quantitative studies. Aside from contextual information, observer, contrast, and the overconfidence effects can impact how forensic scientists interpret evidence. Observer effect is often related to the Hawthorne effect, wherein participants who know that they are being observed tend to do better than without knowledge of observation (Page et al., 2012, p.110). Confirmation bias can happen when the police and colleagues become the “researchers” and odontologists act as the “subjects” (Page et al., 2012, p.110). The existence of the perception of observation can affect how odontologists conduct and interpret their work. As for contrast effect, it refers to the “tendency to shift the judgment standard after repeated exposure to stimuli of a certain threshold and is particularly inherent in subjective comparison work, such as that performed by forensic odontologists” (Page et al., 2012, p.110). Odontologists are vulnerable to contrast effects when they soon “see” the connection “between the mark and the dentition after lengthy analysis” (Page et al., 2012, p.110). Conducting the analysis in relation to suspect dentition can instigate bias through the “target-shifting” method, where the threshold for recognizing the important of the mark is reduced as the analysis goes on (Page et al., 2012, p.110). In other words, odontologists begin to see something that is not actually present in the evidence. Aside from contrast effect, overconfidence effect is another source of bias. It can come from the belief of one’s excellence because of tasks that are done repetitively (Page et al., 2012, p.110). Page et al. (2012) underscored that though confidence is not always related to poor reliability, the confidence of forensic analysts should not be equated to the validity and reliability of their work (p.110). Overconfidence can reduce the effectiveness of forensic experts in treating evidence as new, instead of having assumptions already about it even before tests are conducted. These sources of biases can be present at the workplace of forensic scientists, or may come from their interactions with different stakeholder groups. Another article, however, questioned the limitations of contextual ignorance to the validity of forensic analysis. In “What Role Should Investigative Facts Play in the Evaluation of Scientific Evidence?” Thompson (2011) investigated the suitable role of investigative facts in the study and interpretation of scientific evidence and the proper role of forensic scientists in criminal investigations. He noted that in real life, the CSI TV series do not accurately reflect the range of functions of forensic analysts. For him, analysts are more of lab experts, though they sometimes interact with law enforcement agencies and lawyers. Furthermore, Thompson (2011) argued that forensic evidence is critical to forensic scientists because it helps them to understand the placement of evidence. Context enables them to understand the placement of the evidence in the overall picture of criminal events. Contextual information, hence, is not always improper information for forensic analysts. In addition, Thompson (2011) undermined the role of blind procedures in assessing evidence. He stated: “One common response is that blind procedures are unnecessary for individuals who have proper values and standards of personal integrity. Those who give this response often claim to be insulted at the very suggestion that they might be biased” (Thompson, 2011, p.129). Then again, this paper believes that blind procedures do not have to be an emotional issue for forensic scientists. They should accept it as part of their job, in order to preserve high validity and reliability in their work. Contextual ignorance may not be perfectly desirable, but it can be somewhat attained to avoid the impact of emotional or cognitive bias on forensic analysts’ interpretations. Aside from biases, lack of expertise through training and/or experience can affect forensic scientists’ work outcomes, where expertise may be inversely related to cognitive bias. In the article, “Testing for Potential Contextual Bias Effects during the Verification Stage of the ACE-V Methodology when Conducting Fingerprint Comparisons,” Langenburg, Champod, and Wertheim (2009) studied if fingerprint specialists could be affected by extraneous contextual information during a verification process. Participants were divided into three groups: control, low-bias, and high-bias groups. An identical experiment was done with laypeople. Findings showed that contextual information affected fingerprint experts during fingerprint comparisons, but not in making errors. Fingerprint experts in the biased conditions gave less mistaken conclusions than the control group, and novice participants were affected by bias more and were inclined to offer erroneous judgment. Expertise and training, which can come from experience and background/continuing education, can affect how forensic experts view and analyze evidence. Expertise, however, is not the same across forensic analysts, a problem that is present in many other fields. Its evaluation and continuous development is critical to the reliability and validity of forensic science work. Because of the fallibility of human reasoning, technology can be used to enhance, but not necessarily replace, forensic analysis and interpretation. In “The Use of Technology in Human Expert Domains: Challenges and Risks Arising from the Use of Automated Fingerprint Identification Systems in Forensic Science,” Dror and Mnookin (2010) studied distributed cognition as it is connected to the domain of forensic science: Automated Fingerprint Identification Systems (AFIS). They explored its use, in connection to latent fingerprint identification in the criminal justice system. They learned that AFIS may produce new biases because not all matches are 100% ones, but they may be interpreted as such. Its outcomes can result to contextual information that can prevent forensic scientists from fully evaluating the validity of AFIS interpretation. Dror and Mnookin (2010) concluded that technology can enhance the efficiency and effectiveness of forensic science, but it cannot be accepted as perfect. Human beings and their cognition, including contextual understanding, can yield additional interpretative layers that can improve forensic analysis and interpretation. Bias cannot be completely removed in forensic science, but it can be managed and reduced. Page et al. (2012) suggested the separation of the phases of collection and analyses and decreasing exposure to contextual information for forensic experts who are in charge of analysis. They underscored that odontologists, for instance, should avoid dealing with lawyers, law enforcement agencies, and victims as much as possible to reduce exposure to potentially-biasing sources of biases (Page et al., 2012, p.110). Page et al. (2012) also stressed that poor or ambiguous evidence must be properly categorized and presented as such before and after analysis and interpretation. Bite marks, for instance, are not entirely ideal specimens, and sometimes, they may be corrupted already by other elements. Moreover, odontologists must present the quality of the evidence in relation to the quality of their analyses to provide the judge and jury the opportunity to understand that the evidence may not be entirely conclusive. Honesty, when it comes to presenting interpretation, allows judges/jury members to implement reasonable caution in accepting evidence. Truth is important to the execution of justice, and this includes the truthfulness on the quality of evidence being analyzed. The range of studies in this paper suggested that cognitive bias and other sources of bias affect the work of forensic experts. As human beings, they have their own personal biases and they can be influenced by external conditions and information. The key is not eliminating bias, which is unrealistic and may possibly be unethical, but in managing it properly. The first step is acceptance of its existence, and the subsequent steps are determining existing sources and how they affect analysis and interpretation. From here, studies suggested separation of collection and analysis to reduce cognitive bias and reduction of interaction for analysts with other stakeholders. They need to know some contextual information to understand the evidence better, but they do not need emotive words and labels that can impact their judgment adversely. Future studies have to be conducted on how different fields may have different sources of biases and how they can be determined and managed. Qualitative studies are also needed to understand the human processes involved in resolving bias effects on forensic science. References Bergman, G., & Fox, A. (2009). Lessons from D.A.R.E.: The complicated relationship between research and practice. Center for Court Innovation, 1-12. Cohen, G., & Smith, R. (2010). The racial geography of the federal death penalty. Washington Law Review, 85(3), 425-492. Dror, I.E. (2012). Cognitive forensics and experimental research about bias in forensic casework. Science & Justice, 52(2), 128-130. Dror, I.E., & Mnookin, J.L. (2010). The use of technology in human expert domains: challenges and risks arising from the use of automated fingerprint identification systems in forensic science. Law, Probability & Risk, 9(1), 47-67. Dror, I., & Rosenthal, R. (2008). Meta-analytically quantifying the reliability and biasability of forensic experts. Journal of Forensic Sciences, 53(4), 900-903. Giannelli, P. (2010). Cognitive bias in forensic science. Criminal Justice, 25(2), 61-63, 68. Langenburg, G., Champod, C., & Wertheim, P. (2009). Testing for potential contextual bias effects during the verification stage of the ACE-V Methodology when conducting fingerprint comparisons. Journal of Forensic Sciences, 54(3), 571-582. Page, M., Taylor, J., & Blenkin, M. (2012). Context effects and observer bias-implications for forensic odontology. Journal of Forensic Sciences, 57(1), 108-112. Pinchi, V., Norelli, G., Caputi, F., Fassina, G., Pradella, F., & Vincenti, C. (2012). Dental identification by comparison of antemortem and postmortem dental radiographs: Influence of operator qualifications and cognitive bias. Forensic Science International, 222(1-3), 252-255 Thompson, W.C. (2011). What role should investigative facts play in the evaluation of scientific evidence? Australian Journal of Forensic Sciences, 43(2/3), 123-134. Read More