A basic principle of psychology says that a person’s desires and expectations will influence how they perceive and interpret what they see. Is neutrality then a myth? Will a person always be influenced by what they have observed or are told? More importantly, will this information impact the conclusion they reach? In many aspects of life, bias isn’t a big deal, but when it comes to science (and especially forensic science), where the desired outcome is linking the correct suspect to a crime, the subject of bias draws more attention. Is it possible that an analyst working on a case is experiencing bias (either with or without their knowledge), and if so, what impact does this have on the tests performed and ultimately the conclusions made?
Written by: Tara Luther, Promega Corporation
In the 2002 article, The Daubert/Kumho Implications of Observer Effects in Forensic Science: Hidden Problems of Expectation and Suggestion, D. Michael Risinger, et al. sought to determine whether bias currently existed in forensic science. According to Risinger, et al, at the time they wrote their article, published standards “pay significant attention to preventing contamination of evidence and none to preventing contamination of examiners”. They also note that (as of 2002), there have been no formal studies on the actual practices in forensic science laboratories that would document the level to which “observer effects” is present.
Concerns raised in the article include the subject of “observer effects”; if a person is presented with a hypothesis, they tend to be more likely to find data that supports that hypothesis and may in turn overlook data that is contradictory. The article notes that where the evidence is clear, however, these biases can be overridden.
The article notes that a submission to a crime laboratory may include letters from investigators that include more information than is necessary to perform examinations of the evidence. These letters sometimes include other inculpatory evidence that has been found in the case or what the investigator making the submission expects/hopes the requested tests will conclude. Is it possible that this information could bias the analyst performing tests?
Risinger, et al include many instances in which bias may have influenced the outcome of the case, and ultimately conclude that blind testing should be implemented in the forensic science field. They argue that if an analyst isn’t privy to information that is irrelevant to testing, then they can’t be influenced by it. Yet, as earlier mentioned, no formal studies have been done on the actual practices in forensic science laboratories.
In 2009, The Committee on Identifying the Needs of the Forensic Sciences Community, National Research Council published Strengthening Forensic Science in the United States: A Path Forward. The report covered an abundance of topics related to the Forensic Science field, but on page 183 is a chapter called “Improving Methods, Practice, and Performance in Forensic Science”.
The report echoes much of the concerns raised by Risinger, et al. and states that “The forensic science disciplines are just beginning to become aware of contextual bias and the dangers it poses.” The report goes on to say that such bias may be so subtle that the analyst is not aware of it, yet decisions on what tests need to be performed and in what order could be influenced by this bias, which could potentially skew results. The report also addresses concerns that those forensic scientists who are hired by law enforcement agencies or prosecutors’ offices may be subject to bias, and that bias could be introduced through decisions made during the evidence collection process, which could in turn effect who is listed as a suspect.
The Research Council concludes that “The National Institute of Forensic Science (NIFS) should encourage research programs on human observer bias and sources of human error in forensic examinations. Such programs might include studies to determine the effects of contextual bias in forensic practice (e.g., studies to determine whether and to what extent the results of forensic analyses are influenced by knowledge regarding the background of the suspect and the investigator’s theory of the case)… Based on the results of these studies, and in consultation with its advisory board, NIFS should develop standard operating procedures to minimize, to the greatest extent reasonably possible, potential bias and sources of human error in forensic practice. These standard operating procedures should apply to all forensic analyses that may be used in litigation.”
The topic of bias in forensic science has not disappeared since 2009, and doing a simple Google search provides many articles discussing the possibility. We conclude with sentiment from one analyst with 12 years’ experience at 3 different agencies, who argues that perhaps not all bias should be viewed as negative.
Below is one perspective of an analyst with 12 years of experience at 3 different agencies.
I’d like to talk about the role of personal bias in working forensic cases. There was a kind of a splash in 2009 when the PNAS dropped the article critiquing all the forensic sciences and how they’re practiced in the US, where their recommendation 5 on page 191 states that they believe that scientists should basically operate in a vacuum when analyzing a case and not receive any case history or anything and just process every case the same.
I think that most scientists are not really affected by personal bias on the basis that we process so many cases, it’s really difficult to get personally attached to any one, and in the cases when you actually are viscerally affected by that case narrative, maybe it’s for good reason.
For me, I get upset by cases of ongoing sexual abuse of small children and when victims have been raped and beaten to the extent that they are long-term unconscious and hospitalized. Everything else just gets treated the same on my plate, but I feel like I try to go the extra mile in those cases that get me upset.
Our lab gives us the latitude to choose what samples we want to process and we have the option to run samples on a much slower, but more sensitive genetic analyzer for when you really want to work on some low level data. Maybe we’ll try to upload a more complex CODIS profile that will be more difficult to explain in court, but that we felt was important to develop any kind of lead for the police investigators.
While no victim gets less “justice” than they deserve in our lab, some special victims get a tiny bit more at the cost of greater time involvement on our part. It would be unfeasible to process every sample that goes through the lab with this enhanced treatment and realistically, it doesn’t add that much, but knowing that I’ve exhausted all options when handling a case that gets under my skin helps me be ok with the sheer amount of awfulness that we see on a daily basis.
WOULD YOU LIKE TO SEE MORE ARTICLES LIKE THIS? SUBSCRIBE TO THE ISHI BLOG BELOW!