Cognitive Human Factors and Forensic Document Examiner Methods and Procedures

Here we report initial findings from an interdisciplinary study to empirically explore the reliability, measurement validity, and accuracy of established FDE methods and procedures, and to investigate the influence of possible sources of cognitive bias in the methods and procedures of forensic handwriting examination. This article reports findings of our analysis of the relationship between the position of the known signatures and the utilization of writing features in questioned/known signature comparison tasks. Forty-nine professional forensic document examiners from government labs and private participated in an eye-tracking experiment to investigate patterns in their visual behavior as they performed a series of handwriting comparisons. Eye-tracking results revealed a left-to-right and proximal-to-distal pattern of gaze behavior. Significant differences in mean gaze fixation counts were found according to whether the signatures were text-based or stylized; whether the signatures were high or low complexity; whether the signatures were genuine or non-genuine (ground truth); and whether FDE decisions were accurate or misleading. Binomial logistic regression analyses revealed that ground truth and utilization of the upper left known signature were significant predictors of FDE decision accuracy.