Health Science Reviews

By Michaela Enrich

Do forensic practitioners remain unbiased in their respective practices? Several studies that have been conducted within different forensic fields, have shown that there is an increase in the capacity for bias. Since forensic practitioners rely mostly on interpretations, they can be subject to cognitive bias and result in a loss of objectivity. An interesting phenomenon that has been theorized is that there is this need for closure within the forensic disciplines, which could lead to biased conclusions. The concept of cognitive bias in many disciplines has not been fully assessed or understood. The researchers in this paper designed a pilot study that would examine cognitive bias in forensic anthropologist’s interpretations. This study also set to determine which psychological factors might play a role in causing the bias outcomes and results.

Let’s dive into how this was accomplished

The researchers created three separate websites, each containing identical images of skeletal elements. The contextual information provided on these websites differed. Website A “mass grave” provided contextual information that the images originated from a recent mass grave and human right excavations. Website B “Archaeological/Dorset” provided less contextual information that suggested a low level of trauma. Website C “Control” provided no contextual information and was designed as a control group. The websites were synchronized into one URL address. When the participants accessed the link, website A, B or C was randomly chosen for them. The aim of these websites was not for the participants to indicate the type of trauma, but rather to state the level of trauma. The participants were unaware of the true intentions for the study; they were told it was to examine the effectiveness of analysing trauma from photos. This was done to ensure that the results were not swayed from the knowledge that bias was being examined.

The findings…

The data was analysed using the Kruskal-Wallis test, followed by a Mann-Whitney U-test. The survey showed significant variation in the response to trauma from the three different websites, with website A indicating more responses of “high likelihood” or “definite” trauma indications in general when compared to website B and C. The data implies a correlation between the photo trauma interpretation and the supplied contextual information. The different levels of contextual information provided could have impacted the participants by altering their expectations and motivations. This can lead to participants excluding any evidence that does not align with their preconceived ideas. Another interesting find is that in the images where the trauma was ambiguous (i.e. more susceptible to subjective interpretations), cognitive bias was indicated, with website A indicating higher levels of trauma.

However, the authors note that it is necessary to consider that trauma analyses are difficult by nature, and the survey provided no clarification on the participants experience of trauma analysis. Cognitive bias and subjective errors can never be fully mitigated, but they can be recognised and measured. Therefore, recognising the possible external influences and how to limit these influences during visual trauma examinations should be incorporated into their education and training. While the evidence in this pilot study is very compelling, it does leave many things up for interpretation. Interpretations from ambiguous cases/photos is not really the same as indicating something as traumatic when it clearly is not traumatic. Is the interpretation subjected to cognitive bias or is it the result of subjective interpretation? Should cognitive bias rather be assessed using images of clear trauma/not trauma? This way any effects of cognitive bias will be clearly evident. It is important to note that the trauma analyses were done via photographs, would the results still be the same if the participants could physically analysis the skeletal elements?

Conclusion

That being said, this pilot study still demonstrated that forensic practitioners can be subjected to cognitive bias. More research is necessary to properly examine forensic science for cognitive bias as this study only starts to unravel the perceptual and cognitive processes in visual trauma examination.

Reference

Nakhaeizadeh, S., Hanson, I. & Dozzi, N. 2014. The Power of Contextual Effects in Forensic Anthropology: A Study of Biasability in the Visual Interpretations of Trauma Analysis on Skeletal Remains. Journal of Forensic Sciences. 59(5):1177-1183. DOI:10.1111/1556-4029.12473.