Embodied Conversational Agent-Based Deception Detection: Concealed Information Test using Automated Electrodermal, Vocal, and Oculmentric Measurements

Abstract

An Embodied Conversational Agent-based human deception detection system was tested in a laboratory experiment during which some subjects had committed a mock crime and had guilty knowledge of that crime. Sensor signals were analyzed for their ability to discriminate between guilty/deceptive and innocent/truthful examinees and suitability for conducting automated agent-based deception interviews. Electrodermal activity confirmed the presence of arousal during the agent-based interviews.

Meeting Name

2016 SAI Intelligent Systems Conference, IntelliSys 2016 (2016: Sep. 21-22, London, UK)

Department(s)

Business and Information Technology

Comments

This research was supported by the U.S. Department of Homeland (DHS) Security and the Center for Identification Technology Research (CITeR), a National Science Foundation I/UCRC.

Keywords and Phrases

Agent-based deception interviewing; Credibility assessment; Deception detection; Human risk assessment

International Standard Book Number (ISBN)

978-3-319-56990-1

International Standard Serial Number (ISSN)

2367-3370; 2367-3389

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2018 Springer, All rights reserved.

Publication Date

01 Sep 2018

Share

 
COinS