Improving the Data Interpretation Process in Forensic Fire Debris Analysis

Forensic fire debris analysis involves the examination of evidence from a fire scene to determine if an ignitable liquid is present.  Part of this process includes the subjective recognition and comparison of patterns in data from questioned samples to patterns in data from known reference materials.  Any time a person is tasked with making subjective decisions at key stages of a procedure, they may be susceptible to human factors influences that can impact the reliability of their results.  The proposed research aims to improve the reliability of the data interpretation process in forensic fire debris analysis by investigating the impact of human factors, with a specific focus on the potential for cognitive bias arising from exposure to contextual information. 

First, a ground truth data set will be created using physical samples designed to closely resemble the complexity of realistic fire debris evidence.  Some of these samples will be modified through the addition of gasoline or a similar chemical mixture to obtain a set of data that represent various levels of interpretation difficulty, which will be assessed using an objective, quantitative measure of the data’s degree of support for the identification of gasoline.  Then, practicing forensic fire debris analysts will be invited to interpret data from these ground truth samples as part of an online test designed to determine if and how contextual information influences their judgments and conclusions (Study 1).  Next, a survey of practicing forensic fire debris analysts will be conducted to determine what contextual information they think is task-relevant and at what stage in the analytical process they think it is necessary to know that information (Study 2).  If the results of Study 1 demonstrate that contextual information has a significant influence on the data interpretation process in forensic fire debris analysis, then the information obtained from Studies 1 and 2 will be used to develop practical recommendations and customized human factors mitigation tools tailored specifically for use in this forensic discipline. 

Through these activities, the proposed research will provide insight into the influence of human factors on the data interpretation process in forensic fire debris analysis, increase its foundational body of knowledge, and guide the improvement of its procedural standards.  All of these outcomes will improve the consistency and reliability of forensic fire debris analysts’ results, which will contribute to the fair and impartial administration of justice for all. CA/NCF