What Your Mouse Movements Say About You

Nov. 2, 2018

News
Faculty Research
Keyboard and Mouse

New research out of the Eller College of Management at the University of Arizona finds that if you’re lying, your computer mouse-cursor might let others know.

Eller Professor of Management Information Systems Joe Valacich and his co-authors Jeffry Jenkins at Brigham Young University, Jeff Proudfoot at Bentley University, G. Mark Grimes at the University of Houston and Jay Nunamaker at the Eller College, found that mouse-cursor movements can significantly differentiate between people concealing information and people telling the truth.

jvalacichjpg.jpg

Eller Professor of MIS Joe Valacich
Eller Professor of MIS Joe Valacich

Through an experiment using 66 participants, followed by a specialized screening questionnaire called a Concealed Information Test (CIT)—a polygraph-based questioning technique that has also been used in eye tracking and postural rigidity—the researchers found that when answering baseline and key questions on the screening questionnaire, the response-activation time for people concealing information was slower on key questions than on baseline questions.

“In our experiment, we presented stimuli on a computer screen and had people answer each question using a computer mouse,” says Valacich. “Respondents were required to move the mouse from the bottom middle of the screen to one of the two upper corners of the screen, each of which contained a possible answer. People concealing information were consistently more biased toward the opposite answer.”

In other words, mouse trajectories of people concealing information showed greater attraction toward the opposite response on key questions than on baseline questions or than the mouse trajectories of truthful people.

Further, people concealing information moved the mouse more slowly while responding to key questions, compared with responses to baseline questions or compared with truthful people.

Mouse-cursor movements can be monitored online in people’s natural environments without any specialized hardware or software installed on their computers, which has implications for mass-deployment.

“People who conceal information about adverse behaviors in organizations present a significant danger to both the private and public sectors, and the detection of concealed information is traditionally a difficult, expensive, error-prone and time-consuming task,” says Valacich. “This system can be deployed easily through links in an email and embeds on a website, and data analysis can be computed in under a second.”

Companies can use this type of approach to screen individuals for concealed information and trigger follow-up evaluations as needed or to further screen individuals who are already flagged as potential threats by existing systems, thereby decreasing the number of false positives that may be found through electrodermal activity—which is the measurement of an individual’s emotional responses via electrical conductivity of the skin.

The research is forthcoming in the Journal of the Association for Information Systems under the title “Sleight of Hand: Identifying Concealed Information by Monitoring Mouse-Cursor Movements.”


Header image by Image Catalog, courtesy Flickr.