Abstract
Organizational members who conceal information about adverse behaviors present a substantial risk to that organization. Yet the task of identifying who is concealing information is extremely difficult, expensive, error-prone, and time-consuming. We propose a unique methodology for identifying concealed information: measuring people’s mouse-cursor movements in online screening questionnaires. We theoretically explain how mouse-cursor movements captured during a screening questionnaire differ between people concealing information and truth tellers. We empirically evaluate our hypotheses using an experiment during which people conceal information about a questionable act. While people completed the screening questionnaire, we simultaneously collected mouse-cursor movements and electrodermal activity-the primary sensor used for polygraph examinations-as an additional validation of our methodology. We found that mouse-cursor movements can significantly differentiate between people concealing information and people telling the truth. Mouse-cursor movements can also differentiate between people concealing information and truth tellers on a broader set of comparisons relative to electrodermal activity. Both mouse-cursor movements and electrodermal activity have the potential to identify concealed information, yet mouse-cursor movements yielded significantly fewer false positives. Our results demonstrate that analyzing mouse-cursor movements has promise for identifying concealed information. This methodology can be automated and deployed online for mass screening of individuals in a natural setting without the need for human facilitators. Our approach further demonstrates that mouse-cursor movements can provide insight into the cognitive state of computer users.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1-32 |
| Number of pages | 32 |
| Journal | Journal of the Association of Information Systems |
| Volume | 20 |
| Issue number | 1 |
| DOIs | |
| State | Published - Jan 1 2019 |
Fingerprint
Keywords
- Concealed information
- Deception detection
- Electrodermal activity
- Graded-motor response analysis
- Mouse-cursor movement
- Video Emotion Analysis
ASJC Scopus subject areas
- Information Systems
- Computer Science Applications
Cite this
Sleight of hand : Identifying concealed information by monitoring mouse-cursor movements. / Jenkins, Jeffrey L.; Proudfoot, Jeffrey G.; Valacich, Joseph S; Grimes, G. Mark; Nunamaker, Jay F.
In: Journal of the Association of Information Systems, Vol. 20, No. 1, 01.01.2019, p. 1-32.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Sleight of hand
T2 - Identifying concealed information by monitoring mouse-cursor movements
AU - Jenkins, Jeffrey L.
AU - Proudfoot, Jeffrey G.
AU - Valacich, Joseph S
AU - Grimes, G. Mark
AU - Nunamaker, Jay F
PY - 2019/1/1
Y1 - 2019/1/1
N2 - Organizational members who conceal information about adverse behaviors present a substantial risk to that organization. Yet the task of identifying who is concealing information is extremely difficult, expensive, error-prone, and time-consuming. We propose a unique methodology for identifying concealed information: measuring people’s mouse-cursor movements in online screening questionnaires. We theoretically explain how mouse-cursor movements captured during a screening questionnaire differ between people concealing information and truth tellers. We empirically evaluate our hypotheses using an experiment during which people conceal information about a questionable act. While people completed the screening questionnaire, we simultaneously collected mouse-cursor movements and electrodermal activity-the primary sensor used for polygraph examinations-as an additional validation of our methodology. We found that mouse-cursor movements can significantly differentiate between people concealing information and people telling the truth. Mouse-cursor movements can also differentiate between people concealing information and truth tellers on a broader set of comparisons relative to electrodermal activity. Both mouse-cursor movements and electrodermal activity have the potential to identify concealed information, yet mouse-cursor movements yielded significantly fewer false positives. Our results demonstrate that analyzing mouse-cursor movements has promise for identifying concealed information. This methodology can be automated and deployed online for mass screening of individuals in a natural setting without the need for human facilitators. Our approach further demonstrates that mouse-cursor movements can provide insight into the cognitive state of computer users.
AB - Organizational members who conceal information about adverse behaviors present a substantial risk to that organization. Yet the task of identifying who is concealing information is extremely difficult, expensive, error-prone, and time-consuming. We propose a unique methodology for identifying concealed information: measuring people’s mouse-cursor movements in online screening questionnaires. We theoretically explain how mouse-cursor movements captured during a screening questionnaire differ between people concealing information and truth tellers. We empirically evaluate our hypotheses using an experiment during which people conceal information about a questionable act. While people completed the screening questionnaire, we simultaneously collected mouse-cursor movements and electrodermal activity-the primary sensor used for polygraph examinations-as an additional validation of our methodology. We found that mouse-cursor movements can significantly differentiate between people concealing information and people telling the truth. Mouse-cursor movements can also differentiate between people concealing information and truth tellers on a broader set of comparisons relative to electrodermal activity. Both mouse-cursor movements and electrodermal activity have the potential to identify concealed information, yet mouse-cursor movements yielded significantly fewer false positives. Our results demonstrate that analyzing mouse-cursor movements has promise for identifying concealed information. This methodology can be automated and deployed online for mass screening of individuals in a natural setting without the need for human facilitators. Our approach further demonstrates that mouse-cursor movements can provide insight into the cognitive state of computer users.
KW - Concealed information
KW - Deception detection
KW - Electrodermal activity
KW - Graded-motor response analysis
KW - Mouse-cursor movement
KW - Video Emotion Analysis
UR - http://www.scopus.com/inward/record.url?scp=85067981595&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85067981595&partnerID=8YFLogxK
U2 - 10.17705/1jais.00527
DO - 10.17705/1jais.00527
M3 - Article
AN - SCOPUS:85067981595
VL - 20
SP - 1
EP - 32
JO - Journal of the Association of Information Systems
JF - Journal of the Association of Information Systems
SN - 1536-9323
IS - 1
ER -