Systems and methods for non-intrusive deception detection
Inventors
Macknik, Stephen L. • Martinez-Conde, Susana • DALE, Richard E. • Besserman, Richard • McDaniel, Troy Lee
Assignees
Dignity Health • Arizona State University ASU • Arizona State University Downtown Phoenix campus
Publication Number
US-11759134-B2
Publication Date
2023-09-19
Expiration Date
2035-06-11
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Systems and methods for detecting deceptive intent of a subject include observing eye movements of the subject and correlating the observed movements to known baseline neurophysiological indicators of deception. A detection system may record eye movement data from the subject, compare the eye movement data to a data model comprising threshold eye movement data samples, and from the comparison make a determination whether or not the subject is lying. The detection system may create an alert if deception is detected. The eye movements detected include saccadic and intersaccadic parameters such as intersaccadic drift velocity. Measurements may be collected in situ with a field testing device, such as a non-invasive, non-contact device attached to the subject's computing device and configured to non-obtrusively record the eye movement data.
Core Innovation
The invention provides systems and methods for non-intrusive deception detection by monitoring, recording, and analyzing a subject's eye movements, particularly saccadic and intersaccadic parameters such as intersaccadic drift velocity. A sensing arrangement, which may be integrated into cameras, motion tracking devices, or computing devices, collects real-time eye movement data from a subject. This data is processed and compared to baseline neurophysiological indicators of deception stored in a data model comprising threshold eye movement samples.
The problem addressed is the lack of non-contact, non-invasive, and objective systems capable of detecting deceptive intent without depending on indirect physiological responses (such as those used in polygraphy) or subjective behavioral cues. Current methods tend to require skilled operators, are invasive, or are impractical for remote communication scenarios such as online interactions. The invention aims to protect users of internet services, social media, and other contexts where assessment of truthfulness is challenging due to a lack of face-to-face interaction.
By directly analyzing patterns of oculomotor dynamics that are not under voluntary control, the system determines whether a subject's brain function has altered from a normal state due to deceptive intent. Upon detecting a deviation from baseline measurements (e.g., a current intersaccadic drift velocity below a threshold by more than a threshold amount), the system generates a perceptible alert, notifying relevant parties of potential deception. The innovation allows continuous, in situ monitoring, and can be deployed in wearable devices, mobile devices, or over social networks, providing objective and automated deception detection.
Claims Coverage
There are four independent claims in the patent, each directed to inventive features for systems and methods of deception detection using eye movement analysis.
System for detecting deceptive intent using eye movement dynamics
A system comprises a sensing arrangement to collect a user's eye movement data, an alerting arrangement to produce a perceptible alert, and a control unit configured to: - Extract current eye movement dynamics, including intersaccadic drift velocity. - Compare current dynamics to baseline eye movement dynamics that include a threshold drift velocity. - Identify deceptive intent when the current intersaccadic drift velocity falls below the threshold by more than a threshold amount. - Send an alert signal to the alerting arrangement upon identifying deception.
Method for determining whether a subject is lying based on eye movement drift velocity
A method includes: 1. Obtaining eye movement data from one or both eyes of a subject. 2. Identifying a current intersaccadic drift velocity from the data. 3. Using a control unit to compare the current drift velocity to a threshold drift velocity. 4. When the current drift velocity is below the threshold by more than a threshold amount, indicating that the subject is lying, delivering an alert to a device associated with the subject or an administrator. - The threshold drift velocity is part of one or more bio-signatures corresponding to response conditions, including deceptive answers.
Method for determining whether a subject is lying using baseline eye movement data and threshold calculation
A method involves: 1. Obtaining baseline eye movement data during known deceptive or non-deceptive conditions. 2. Calculating a threshold drift velocity from the baseline data. 3. Obtaining eye movement data of the subject. 4. Identifying the subject's current intersaccadic drift velocity. 5. Comparing the current drift velocity to the threshold. 6. When the current drift velocity is below the threshold by more than a threshold amount, indicating deception, delivering an alert to a device associated with the subject or an administrator.
Method for determining whether a subject is lying with response-specific drift velocity analysis
A method that includes: - Obtaining eye movement data of one or both eyes. - Identifying a current intersaccadic drift velocity from the data, particularly when the subject responds to a question. - Comparing the current drift velocity to a threshold drift velocity using a control unit. - If the current drift velocity is below the threshold by more than a threshold amount, indicating lying, delivering an alert to a device associated with the subject or an administrator.
The independent claims collectively protect systems and automated methods for non-intrusive deception detection by analyzing intersaccadic drift velocity and other eye movement dynamics, comparing them to baseline or threshold values, and providing alerts when deception is detected.
Stated Advantages
Enables non-invasive, non-contact detection of deception based on direct measurement of oculomotor dynamics, rather than indirect physiological responses.
Applicable for remote and online interactions, offering a practical solution for environments where face-to-face assessment is not possible.
Does not require extensive training to operate, reducing reliance on expert operators and subjective interpretation.
Provides objective, automated detection that is more resistant to attempts to defeat the testing process through physiological countermeasures.
Documented Applications
Screening tool for internet and social media communications where users rely on the truthfulness of others, such as online dating, introductory services, and financial transactions.
Used during phone calls, video chats, emails, or text-based messaging to alert recipients when deception is detected.
Integration into devices such as eyewear, mobile phones, tablets, and personal computing devices for real-time, in situ monitoring.
Applicable in law enforcement and security clearance processes as an alternative or supplement to polygraph tests.
Interested in licensing this patent?