Enhancing text-based analysis using neurophysiological measures

Abstract

Intelligence analysts are faced with the demanding task of identifying patterns in large volumes of complex, textual sources and predicting possible outcomes based on perceived patterns. To address this need, the Advanced Neurophysiology for Intelligence Text Analysis (ANITA) system is being developed to provide a real-time analysis system using EEG to monitor analysts' processing of textual data during evidence gathering. Both conscious and unconscious 'interest' are identified by the neurophysiological sensors based on the analyst's mental model, as related to specific sentences, indicating relevance to the analysis goal. By monitoring the evidence gathering process through neurophysiological sensors and implementation of real-time strategies, more accurate and efficient extraction of evidence may be achieved. This paper outlines an experiment that focused on identifying distinct changes in EEG signals that can be used to decipher sentences of relevance versus those of irrelevance to a given proposition. © 2009 Springer.