Psychophysical Evaluation of Human Performance in Detecting Digital Face Image Manipulations
Open accessThe DOI matches its Crossref record; 71 of the 71 references it cites were found in a registry or an index.
Abstract
In recent years, increasing deployment of face recognition technology in security-critical settings, such as border control or law enforcement, has led to considerable interest in the vulnerability of face recognition systems to attacks utilising legitimate documents, which are issued on the basis of digitally manipulated face images. As automated manipulation and attack detection remains a challenging task, conventional processes with human inspectors performing identity verification remain indispensable. These circumstances merit a closer investigation of human capabilities in detecting manipulated face images, as previous work in this field is sparse and often concentrated only on specific scenarios and biometric characteristics. This work introduces a web-based, remote visual discrimination experiment on the basis of principles adopted from the field of psychophysics and subsequently discusses interdisciplinary opportunities with the aim of examining human proficiency in detecting different types of digitally manipulated face images, specifically face swapping, morphing, and retouching. In addition to analysing appropriate performance measures, a possible metric of detectability is explored. Experimental data of 306 probands indicate that detection performance is widely distributed across the population and detection of certain types of face image manipulations is much more challenging than others.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Overall, deepfake detection does not seem to vary based on gender, education, income, or profession [7, 32, 37].”
From Human Performance in Deepfake Detection: A Systematic Review (2025) · cites this paper as [7]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Type of stimuli (video versus audio) and the technique used for deepfake generation (e.g., face swapping, lip-syncing, and level of morphing) demonstrated mixed results [7, 20, 22, 28, 30, 42].”
From Human Performance in Deepfake Detection: A Systematic Review (2025) · cites this paper as [7]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Type of stimuli (video versus audio) and the technique used for deepfake generation (e.g., face swapping, lip-syncing, and level of morphing) demonstrated mixed results [7, 20, 22, 28, 30, 42].”
From Human Performance in Deepfake Detection: A Systematic Review (2025) · cites this paper as [7]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Furthermore, Nichols et al. found that participants were more likely to make false-negative errors (i.e., saying a photo is fake, when it is real) for lower-quality morphs. The inverse was true for higher-quality morphs -participants were more likely to make false-positive errors (i.e., saying a photo is real, when it is fake).”
From Human Performance in Deepfake Detection: A Systematic Review (2025) · cites this paper as [7]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Almost all video/image stimuli used face swapping or lip-sync technique, except for three studies that used morphing techniques [6, 7, 18].”
From Human Performance in Deepfake Detection: A Systematic Review (2025) · cites this paper as [7]
5 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
Deng, J., Guo, J., Xue, N., & Zafeiriou, S. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00482 · Cited by 7,781 · cited 1×
“Similarity scores were computed for each manipulated face image from a comparison with the bona fide source image, by means of ArcFace face recognition system (FRS), based on a deep convolutional neural network (DCNN) approach utilizing Additive Angular Margin Loss [80], for”
MethodSensory Evaluation of Food: Principles and Practices
Lawless, H. T., & Heymann, H. (2010) · Food Science Text Series · doi:10.1007/978-1-4419-6488-5 · Cited by 4,815
FaceForensics++: Learning to Detect Manipulated Facial Images
Rossler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., & Niessner, M. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00009 · Cited by 3,314 · cited 5×
“One of few studies examining alternative manipulation methods is [54], where the human ability to distinguish deepfake (i.e. face swapping) images from originals was assessed.”
BackgroundMethodResultInfluentialCalculation of signal detection theory measures
Stanislaw, H., & Todorov, N. (1999) · Behavior Research Methods, Instruments, & Computers · doi:10.3758/bf03207704 · Cited by 2,673
Overview of the Face Recognition Grand Challenge
Phillips, P. J., Flynn, P. J., Scruggs, T., Bowyer, K. W., Chang, J., Hoffman, K., Marques, J., Min, J., & Worek, W. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.268 · Cited by 2,438
The FERET database and evaluation procedure for face-recognition algorithms
Phillips, P. J., Wechsler, H., Huang, J., & Rauss, P. J. (1998) · Image and Vision Computing · doi:10.1016/s0262-8856(97)00070-x · Cited by 2,431
The psychometric function: I. Fitting, sampling, and goodness of fit
Wichmann, F. A., & Hill, N. J. (2001) · Perception & Psychophysics · doi:10.3758/bf03194544 · Cited by 2,358
First Impressions: Making Up Your Mind After a 100-Ms Exposure to a Face
Willis, J., & Todorov, A. (2006) · Psychological Science · doi:10.1111/j.1467-9280.2006.01750.x · Cited by 2,300 · cited 1×
“Faces in particular represent a group of stimuli which have been found to be processed very quickly; in terms of identity representations as fast as 30 milliseconds [74], [75].”
BackgroundJ., J., & Fechner, G. T. (1889) · The American Journal of Psychology · doi:10.2307/1411906 · Cited by 1,429
Macmillan, N. A., & Creelman, C. D. (2004) · doi:10.4324/9781410611147 · Cited by 1,269
Research works printed but not linked: 12
- [1] W. W. Bledsoe, The Model Method in Facial Recognition , vol. 15. Palo Alto, CA, USA: Panoramic Research, 1966, p. 2. -
- [8] D. M. Green and J. A. Swets, Signal Detection Theory Psychophysics . New York, NY, USA: Wiley, 1966. -
- [19] F. Richter. (Jul. 2019). Faceapp Craze Reaches New Heights . [Online]. Available: https://www.statista.com/chart/18769/estimated-worldwidefaceapp-downloads-by-platform/ -
- [20] Airnow. (Jul. 2021) Leading Entertainment APP Titles in the Apple App Store in Spain . [Online]. Available: https://www.statista. com/statistics/699448/leading-iphone-entertainment-apps-in-spain-bydownloads/ -
- [26] G. Wolberg, Digital Image Warping , vol. 10662. Los Alamitos, CA, USA: IEEE Computer Society Press, 1990. -
- [34] A. Boyd. (Apr. 2021). CBP's Facial Biometrics Program has Caught 26 Alleged Imposters . [Online]. Available: https://www.defenseone.com/technology/2018/11/cbps-facial-biometricsprogram-has-caught-26-alleged-imposters/152993/ -
- [36] Information Technology Biometric Presentation Attack Detection Part 1: Framework , Standard ISO/IEC 30107-1, International Standards Organization, 2016. -
- [39] FastPass. (Aug. 2021). A Harmonized, Modular Reference System for all European Automated Border Crossing Points . [Online]. Available: https://www.fastpass-project.eu/ -
- [64] J. M. Ennis, B. Rousseau, and D. M. Ennis, ''Proper task instructions and the two-out-of- ve-test,'' IFPress , vol. 15, no. 4, pp. 3 4, 2012. -
- [69] Information Technology Biometric Presentation Attack Detection Part 3: Testing and Reporting , Standard ISO/IEC 30107-3, International Standards Organization, 2017. -
- [78] Fotoable. (2021). Fotorus Camera & Photo Editor . [Online]. Available: https://apps.apple.com/us/app/fotorus-camera-photo-editor/id457517348 -
- [79] Fotoable. (2021). Instabeauty Makeup Camera . [Online]. Available: https://apps.apple.com/us/app/instabeauty-makeup-camera/id599534650 -
Cited by
Deepfakes as a threat to a speaker and facial recognition: An overview of tools and attack vectors
Firc, A., Malinka, K., & Hanáček, P. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e15090 · Cited by 80
Human Performance in Deepfake Detection: A Systematic Review
Somoray, K., Miller, D. J., & Holmes, M. (2025) · Human Behavior and Emerging Technologies · doi:10.1155/hbe2/1833228 · Cited by 19
Challenges for automated face recognition systems
Busch, C. (2024) · Nature Reviews Electrical Engineering · doi:10.1038/s44287-024-00094-x · Cited by 13
Self-citationOverview of Research Progress of Digital Image Processing Technology
Huang, Y. (2022) · Journal of Physics Conference Series · doi:10.1088/1742-6596/2386/1/012034 · Cited by 11
Crowd-Powered Face Manipulation Detection: Fusing Human Examiner Decisions
Rathgeb, C., Nichols, R., Ibsen, M., Drozdowski, P., & Busch, C. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9897841 · Cited by 10
Self-citationTetraLoss: Improving the Robustness of Face Recognition Against Morphing Attacks
Ibsen, M., Gonzalez-Soler, L. J., Rathgeb, C., & Busch, C. (2024) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg59268.2024.10581988 · Cited by 10
Self-citationConditional Face Image Manipulation Detection: Combining Algorithm and Human Examiner Decisions
Ibsen, M., Nichols, R., Rathgeb, C., Robertson, D. J., Davis, J. P., Løvåsdal, F., Raja, K. B., Jenkins, R. E., & Busch, C. (2024) · ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec) · doi:10.1145/3658664.3659649 · Cited by 7
Self-citationThe Super‐Recogniser Advantage Extends to the Detection of Digitally Manipulated Faces
Davis, J. P., Robertson, D. J., Jenkins, R. E., Ibsen, M., Nichols, R., Babbs, M., Rathgeb, C., Løvåsdal, F., Raja, K. B., & Busch, C. (2025) · Applied Cognitive Psychology · doi:10.1002/acp.70053 · Cited by 6
Self-citationHu, T., Fu, Z., Wang, N., Gui, J., Song, Q., & Qian, X. (2025) · Education and Information Technologies · doi:10.1007/s10639-025-13541-5 · Cited by 5
On the Human Ability in Detecting Digitally Manipulated Face Images
Franco, A., Løvåsdal, F., & Maltoni, D. (2023) · IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) · doi:10.1109/metroxraine58569.2023.10405682 · Cited by 4
The super-recogniser advantage extends to the detection of digitally manipulated facesPreprint
Davis, J. P., Robertson, D. J., Jenkins, R. E., Ibsen, M., Nichols, R., Babbs, M., Rathgeb, C., Løvåsdal, F., Raja, K. B., & Busch, C. (2024) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/ye7ph · Cited by 3
Self-citationBiometric Template Protection: Why and How
Busch, C., Gomez‐Barrero, M., & Shahreza, H. O. (2025) · Springer eBooks · doi:10.1007/978-981-95-0995-9_1 · Cited by 2
Self-citationZeng, Z., Ishmael, D., & Lv, Y. (2023) · Forensic Sciences Research · doi:10.1093/fsr/owad041 · Cited by 2
Morphing Resilient Face Recognition by Informed Frequency Selection
Huber, M., Luu, A. T., & Damer, N. (2025) · doi:10.1145/3746265.3759671 · Cited by 1
Diagnosing Progressive Face Recognition from Face Morphing Using ViT Technique Through DL Approach
Faizal, M. K. M., Geetha, S., & Barveen, A. (2023) · International Conference on Networking and Communications (ICNWC) · doi:10.1109/icnwc57852.2023.10127374 · Cited by 1
Human-AI Ensembles Improve Deepfake Detection in Low-to-Medium Quality VideosPreprint
Postiglione, M., Gortner, I., & Subrahmanian, V. S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.14658 · Cited by 1
Study of Deep Learning Algorithm for Uncovering Deep Fakes in Online Networking Platform
Manimozhi, I., Padmavathy, R., Liz, D. R. A. S., Devi, M., Rajkumar, D., & Kavitha, P. (2024) · International Conference on Power, Energy, Control and Transmission Systems (ICPECTS) · doi:10.1109/icpects62210.2024.10780362
The super-recogniser advantage extends to the detection of digitally manipulated facesPreprint
Davis, J. P., Robertson, D. J., Jenkins, R. E., Ibsen, M., Nichols, R., Babbs, M., Rathgeb, C., Løvåsdal, F., Raja, K. B., & Busch, C. (2025) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/ye7ph_v2
Self-citationAlzghoul, E. A. (2024) · UNSWorks (UNSW Sydney) · doi:10.26190/unsworks/30611
Group Decision Performance of Human Examiners for Identity-aware Face Image Manipulation Detection
Real, A. E., Nichols, R. C., Busch, C., & Rathgeb, C. (2026) · doi:10.1145/3785318.3785322
Self-citationDigital Face Manipulation Detection in Virtual Reality: Performance and Gaze Behaviour Analysis
Palaz, B., Gürbüz, K., Nichols, R., & Rathgeb, C. (2026) · IEEE International Conference on Artificial Intelligence and eXtended and Virtual Reality (AIxVR) · doi:10.1109/aixvr67263.2026.00030
Self-citationDikananda, F., Nurhakim, B., Kurnia, D. A., Rifai, A., & Praseptiawan, M. (2026) · TEMATIK · doi:10.38204/tematik.v13i1.2996
Dörsch, A., Nichols, R., Busch, C., & Rathgeb, C. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1834376
Self-citationThe Impact of Contextual Information in Social Media Posts on Human Deepfake Detection Accuracy
Heikkilä, S. (2026) · UTUPub (University of Turku)
Metadata
- Authors
- Nichols, R., Rathgeb, C., Drozdowski, P., & Busch, C.
- Affiliations
- Darmstadt University of Applied Sciences (DE)
- Venue
- IEEE Access
- Published
- 2022
- Volume
- 10
- Pages
- 31359-31376
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4226373926
- Cited by
- 16
- References
- 65
- Citations by year
- 2026: 3 · 2025: 5 · 2024: 2 · 2023: 4 · 2022: 2
- FWCI
- 1.06
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Funders
- German Federal Ministry of Education and Research
- Hessian State Ministry for Higher Education, Research and the Arts within their joint support of the National Research Center for Applied Cybersecurity ATHENE
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
- Keywords
- face swapping, face morphing, retouching, identity verification, biometric attacks, face recognition security, psychophysics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.