User modeling for detecting faking-good intent in online personality questionnaires in the wild based on mouse dynamics

• Published in: Multimedia tools and applications Vol. 84; no. 34; pp. 43395 - 43431
• Main Authors: Kuric, Eduard, Demcak, Peter, Smrecek, Peter, Spilakova, Beata
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.10.2025; Springer Nature B.V
• Subjects: Computer Communication Networks; Computer Science; Data Structures and Information Theory; Datasets; Dynamics; Insurance claims; Laboratories; Likert scale; Lying; Machine learning; Multimedia Information Systems; Performance evaluation; Personality; Personality tests; Questionnaires; Special Purpose and Application-Based Systems; Track 9: Machine Learning in Multimedia; User behavior; Big Five personality traits; User modeling; Intent to lie; User behavior analysis; Classification; Mouse dynamics
• ISSN: 1573-7721; 1380-7501; 1573-7721
• DOI: 10.1007/s11042-025-20852-9

With widespread use of online forms and questionnaires, detection of the user’s intent to lie has become increasingly important. In-lab studies have shown that mouse dynamics-information on how the user operates a mouse-can be valuable for automatically and unobtrusively distinguishing between faking and honest intent in personality questionnaires. However, data variability in the wild (e.g., mouse configurations, screen resolutions) could present a barrier for mouse dynamics in naturalistic conditions. Our aim is to design a data-driven faking good detection method operable under naturalistic conditions and evaluate its reliability. We conducted a between-subjects uncontrolled experiment where users’ mouse dynamics data was obtained with the BFI-2 personality questionnaire, yielding a sample of 5344 items from 112 participants. The proposed user model characterizes participants’ mouse dynamics, serving as input for prediction. Varied machine learning classification models were trained and evaluated to determine the best approach and feature set. XGBoost with LASSO feature selection achieved the highest F1-score of 86.92%, outperforming in a real-world web form and conditions the results achieved by related works utilizing specialized questions in laboratory conditions. Faking good prediction that leverages our user model can therefore be viable for online questionnaires in the wild.