Priming in fire pattern intepretation

Abstract

Fire investigator’s task is to find out the cause of the fire. One step in determining the origin and cause of the is the fire pattern interpretation, in which the fire investigator examines the damage caused to the materials in the fire to determine the direction of fire spread and the origin of the fire. Because most of the analysis of data collected is based on the investigator’s personal interpretation fire investigation can be considered particularly vulnerable to the negative effects of decision-making phenomenon.

Fire investigation is one of the forensic sciences. Forensic decision making is influenced by many different factors, such as case-related information, policy and organizational factors, and cognitive factors. One of these potentially influencing cognitive factors is priming, which means the effect of the first stimulus, the prime, on a later activity or decision. Priming can be done, for example, verbally, visually or by writing and it can occur either consciously or subconsciously, but the object of the priming must not recognize the expected effect of the priming.

The purpose of this thesis was to produce novel information about the effect of priming in fire investigation. The research question was that how does priming affect fire pattern intepretation. The thesis was carried out with a quantitative research approach because the purpose is to produce measurable information about an individual phenomenon. The study was conducted as a randomized experiment. This method was chosen because with other methods the effect of priming would have been very difficult to distinguish.

Fire investigators from six different countries working in the police, rescue services and other organizations participated in the experiment between 14.10.2019 and 23.2.2020. Participants were randomized in one of three groups: intentional fire priming, electric fire priming, and control group. The randomization method was simple randomization. The priming was produced by instructing the participant to recall a previously investigated fire and write a brief description about it. The participant was then asked to perform two fire pattern intepretation tasks based on photographs, and to evaluate the confidence of their intepretation on a scale of 0–10. The responses of 79 fire investigators were accepted for analysis. The data was analysed using multiple different statistical analysis methods.

Based on this study priming can affect fire pattern intepretation. The priming of an incendiary fire reduced the certainty of interpretation of an irregular fire pattern in a statistically significant manner compared to the control group. The result can be explained by loss avoidance or the activation of a scheme related to intentionally set fires. The results of the thesis can be used to increase awareness of the decision making phenomenon and to improve the fire investigation process. Fire investigators are instructed to use their experience in data analysis but based on the results they should avoid involvement to old cases during the data analysis phase of a current investigation. The results are also applicable to other domains were past work cases are used in visual identification tasks, such as medical decision making and mechanical failure analysis.