The Visualisation of Cognitive Structures in Forensic Statements

Forensic statements are often unstructured, intricate, and thus difficult to interpret and assess. This is due to the varied nature and format of how interviews with victims, witnesses, or suspects are conducted. It is even more difficult for police investigators, lawyers or other legal practitioners to grasp intuitively and accurately the key information pertaining within the varied statements. This research investigates the opportunities in the convergence of linguistic approaches to extracting and reconstructing the cognitive structure, i.e. "Text-Worlds", in a statement, and the computerised operational settings for enabling effective and hopefully more accurate interpretation of forensic discourse through visualisation.

1. Introduction

Forensic statements are the text records made by victims, witnesses and suspects. Those statements, along with other physical evidence, are often used to reconstruct truthful crime stories during criminal investigations and court proceedings. However, legal practitioners usually have to face necessarily detailed, extremely complicated and large amounts of text records. It is often challenging for them to identify connections between disparate pieces of linguistic evidence, and to properly assess their probative values to the case in question. Various approaches have been developed for evidence representation, manipulation and automated analysis [1]–[3]. However, a rigorous method that can facilitate the interpretation of detailed forensic statements and assist in objective judgments is still underdeveloped.