Improvement of discrimination method utilizing effect sizes to identify differential physiological responses in the Concealed Information Test

Abstract

 The concealed information test (CIT) is a memory detection technique that examines whether an examinee knows crime-relevant information. In Japanese field applications, an expert judge whether the examinee recognizes crime-relevant information by visually inspecting differences in physiological responses to crime-relevant and crime-irrelevant items. This study proposes an improvement of the statistical method for discriminating differential physiological responses that integrates multiple autonomic measures. The proposed method involves (a) weighting of effect sizes based on the consistency of differential physiological responses and (b) utilizing information on the covariate relationships of physiological indices. This method was applied to CIT data obtained from 167 participants. The proposed method exhibited adequate specificity (95%) and inconclusive rate (20%) for utilization in criminal investigations, though the sensitivity (75%) was moderate. The proposed method is expected to improve the validity of the CIT by increasing sensitivity and specificity through its combination with expert judgement.