Software, Demos, and Prototypes

Automated Screening Kiosk (ASK)

The ASK is a modular kiosk system for human screening for human credibility risk assessment. The ASK features an automated protocol, conducting a standardized interview that controls for many cultural, demographic, and question type effects. Practical risk assessment applications for the ASK range from insider threat detection to pre-employment screening to pre-audit fraud analysis. The ASK hardware platform is configurable, allowing for rapid experimentation of sensor arrays. The ASK’s broad array of sensors and malleable design assists with discovery the contextualized interrelationships among physiological and behavioral indicators of deception.

Experiments using the ASK system have examined feasibility and interrelationship among oculometric, vocalic, kinesic, cardiorespiratory, galvanometric, and mousing activity during conditions of low veracity. Interviewing paradigms examined or scheduled to be examined include variants of the Concealed Information Test (CIT), Behavioral Analysis Interview (BAI), Control Question Technique (CQT), and novel paradigms.

AVATAR Kiosk for Deception Detection

The AVATAR kiosk is an automated interviewing platform with an embedded artificial agent that is designed to flag suspicious behavior at a port-of-entry that should be investigated more closely by a trained officer.  This primary screening technology is designed for use at ports-of-entry, including border crossings and airports.  The kiosk also has many other security application such as visa processing and personnel screening. 

Biometric Camera for Remote Surveillance and Identification

This project uses camera networks that could be suitable for an airport (or other port of entry) security situation.  In the demonstration, a subject will walk in-between cameras which will accumulate videos, collectively select frames where the face is most visible, and submit only the most promising images to facial recognition software.

Checkpoint Simulation

The Checkpoint simulation model was developed for Border Patrol using Arena software. The simulation assists Border Patrol in resource allocation planning. Using the simulation, Border Patrol can analyze current and expected traffic flows to determine how many resources are needed to reach its goals. The simulation provides a graphical view of Checkpoint wait times, as well as detailed statistics on key Checkpoint metrics such as flushing and apprehensions. Using this tool, Border Patrol can quickly answer "what if" questions about changes at Checkpoints.

Dynamic Embodied Agent for Persuasion

Dynamic Embodied Agent for Persuasion (DEAP) is a comprehensive framework which leverages state-of-the-art software to generate, animate and control embodied agents (EA) to interact with users in a variety of settings.  DEAP has real-time text-to-speech capabilities which will allow it to analyze the verbal responses it receives from its interactants.  The EA's speech is driven by text.  A text-to-speech (TTS) engine is used to translate the text to the spoken word and sync the EA's lips with the speech.  This means the EA's dialog can be changed quickly to respond to changes in its environment.  The framework is also capable of switching to and from different EA's in order to best meet the needs of the current circumstance.  For instance, people are often more comfortable with EA of the same ethnicity.  In addition, our framework has the capability to present EAs with similar facial features as the interactant.  Though it still needs to be automated, this is a fascinating aspect of the DEAP framework.

The DEAP framework has been leveraged to run a number of experiment to study the persuasive ability of EAs.  In one study, the EA's attractiveness and argument quality  was found to have a significant effect when trying to influence users' final decision when making credibility assessments.  In another study, the EAs language power was influential in persuade the users' assessments.

Eye Tracker for Vigilance

Screeners at ports of entry (land, sea, air) must process millions of passengers daily. A potential security risk in this system is fluctuating levels of vigilance and attentiveness on the part of screeners monitoring for contraband, forged documents, etc. The purpose of this project is to demonstrate how eye movements can be monitored and interpreted to measure an individual’s awareness and attention while performing this task. The eye tracker demo will show a system designed to monitor and improve vigilance based on pupillometry and blink rates.

Kinect for Deception Detection

The Kinect (developed by Microsoft) enables us to monitor and display the body position of a person standing in front of a sensor. The demo will show the real-time tracking and modeling of body position and motion which has applications for deception detection research. For example, an individual’s defensive posture (e.g., folded arms) could be indicative of deception, while an open posture may indicate truth telling. Combined with the other sensors in the AVATAR kiosk, the Kinect’s models could improve the accuracy of deception classification.

Micro UAV Sensor Network in Remote Desert Regions

A pilot micro UAV-based field test conducted in Tucson demonstrated the ability to transmit a person’s  physiological data (blood pressure, pulse rate, etc.) from a rugged remote terrain back to a central location for monitoring over an ad-hoc communication network.  The GPS-empowered ad-hoc network would also provide a strong backbone through which other real-time border information could be transferred and collected. This demo will show videos of the field tests and display the Micro UAVs used to transmit the data. 

SPLICE

Structured Programming for Linguistic Cue Extraction (SPLICE) is a linguistic analysis tool. It is available as a web service to aid in the development of applications that need to take advantage of powerful natural language processing techniques. There is also a GUI available. SPLICE is currently in Beta testing.

Visit the SPLICE website.

StrikeCOM

StrikeCOM is a multi-player online strategy game useful for researching group collaboration. StrikeCOM is designed to foster discourse among group members and has been used as a research tool to study leadership and deception in group decision making. Researchers use this tool to examine the development of group processes, shared awareness, and communication. The groups can be face-to-face or distributed, allowing researchers to investigate how proximity and synchronicity affects group dynamics and group performance.

The game mimics C3ISR (Command, Control, Communication, Intelligence, Surveillance, Reconnaissance) scenarios and information gathering in group activities. Built using Java and a SQL-based collaborative server platform, the game is available for use in almost any computing environment. The U.S. Department of Defense is using the tool to teach Network Centric Warfare to battle commanders. Use of StrikeCOM over the years has resulted in a number of lessons-learned, including using simple, familiar game interfaces, utilizing full and immediate feedback, and creating a flexible technical design to meet shifting research and teaching needs.

The scenario of the game is for group members to collaborate to discover where an enemy is hiding its missiles. Each user sees a topological map and can allocate surveillance and intelligence resources to search for the hidden missiles. The information gathered by one user can be communicated to his/her teammates. Variations of the scenarios include placing a confederate on the team whose secret task is to mislead the team. This scenario allows researchers to study deception and how groups cope. Other scenarios call for the group members to communicate verbally or by instant messaging to research the effects these restrictions have on group dynamics and group performance. This scenario mimics the real-world where groups may be required to collaborate across distance and time using email or instant messaging.

CMI created the StrikeCOM game to meet requirements for advanced data collection and customizability to investigate deception detection within large groups of people. StrikeCOM is designed for flexibility and allows research into any group process or interaction. StrikeCOM provides a multiplayer game capable of supporting any number of players using any number of assets searching a game board of any definable size for any number of targets and target types that the researcher chooses to define. The game includes time stamps for each player interaction with the game. In addition, the game includes map overlays with adjustable geographic information reliability statistics, the ability to have multiple independent target types, and an optional shared results visualization scheme. The game also allows the researcher to configure the game such that any or all of these features can change as the game progresses.

StrikeCOM Research Findings

  • Twitchell, D. P., Wiers, K., Adkins, M., Burgoon, J. K., & Nunamaker, J. F., Jr. (2005, January 3-6). StrikeCOM: A multi-player online strategy game for researching and teaching group dynamics. Proceedings of the 38th Annual Hawaii International Conference on System Sciences (CD-ROM), Waikoloa, HI.

Pages