Modelling the effect of deception on investigations using open source intelligence (OSINT)

Purpose: A key problem in money laundering investigations based on open source intelligence gathering is the performance (efficiency and accuracy) of people in the team, where suspects will typically try to conceal incriminating evidence or deliberately deceive investigators to avoid prosecution. If we are able to develop a suitable psychological model of deception for web‐based investigations, it should be possible to develop training programmes to improve investigators' ability to “see beyond” deliberate concealment. The purpose of this paper is to empirically test a model based on non‐linear system identification using a well‐known psychological phenomenon (the Stroop effect), where conflicting colour and text information is presented to subjects which they are instructed to process in a certain way.

Design/methodology/approach: The paper uses an experimental approach.

Findings: The results indicate that strategies for improving investigator information processing performance can benefit from models that incorporate both linear and non‐linear components.

Originality/value: Although the Stroop effect is well known, no other papers have investigated how it may be used to evaluate and monitor the performance of investigators. The real value of this study will emerge when tools are developed to better train investigators to identify concealment within conflicting input data.