Human Performance in Deepfake Detection: A Systematic Review
Open accessThe DOI matches its Crossref record; 60 of the 60 references it cites were found in a registry or an index; 79 of the 153 sources its claims cite were checked for support.
Abstract
Deepfakes refer to a wide range of computer‐generated synthetic media, in which a person’s appearance or likeness is altered to resemble that of another. This systematic review is aimed at providing an overview of the existing research into people’s ability to detect deepfakes. Five databases (IEEE, ProQuest, PubMed, Web of Science, and Scopus) were searched up to December 2023. Studies were included if they (1) were an original study; (2) were reported in English; (3) examined people’s detection of deepfakes; (4) examined the influence of an intervention, strategy, or variable on deepfake detection; and (5) reported relevant data needed to evaluate detection accuracy. Forty independent studies from 30 unique records were included in the review. Results were narratively summarized, with key findings organized based on the review’s research questions. Studies used different performance measures, making it difficult to compare results across the literature. Detection accuracy varies widely, with some studies showing humans outperforming AI models and others indicating the opposite. Detection performance is also influenced by person‐level (e.g., cognitive ability, analytical thinking) and stimuli‐level factors (e.g., quality of deepfake, familiarity with the subject). Interventions to improve people’s deepfake detection yielded mixed results. Humans and AI‐based detection models focus on different aspects when detecting, suggesting a potential for human–AI collaboration. The findings highlight the complex interplay of factors influencing human deepfake detection and the need for further research to develop effective strategies for deepfake detection.
Overview
CoChat Summary
TL;DR
This systematic review synthesizes 40 independent studies from 30 records on human deepfake detection, finding widely variable accuracy and mixed effects of person-level, stimuli-level, and intervention factors. Humans sometimes outperform AI models and sometimes underperform them, and most tested interventions were ineffective except interactive/media literacy training and AI prediction feedback.
Key findings
- Across 30 records and 40 independent studies, sample sizes ranged from 19 to 11,088; direct detection accuracy averaged 57.6% to 75.43%, while indirect detection rates ranged from 0% to 83.5%. Metrics varied and included forced choices, Likert scales, open-ended questions, AUC, F1 score, precision, true positive rates, false positive rates, and false negative rates.
- Person-level factors: cognitive ability and analytical thinking were positively associated with detection accuracy; truth bias was associated with miscategorizing deepfakes as real; gender, education, income, and profession had no effect; age and ethnicity showed mixed results. Stimuli-level factors: familiarity with the deepfaked subject improved detection; higher-quality deepfakes were harder to detect; face swaps were easier to detect than morphs; lip-syncs were harder to detect than face-swapped videos; audio deepfakes were poorly detected.
- Human versus AI performance was mixed: Groh et al. found 82% of participants who saw at least 10 videos outperformed the winning DFDC model, and the crowd mean identified 74%–80% of videos versus the model’s 80%; Korshunov and Marcel found humans achieved average AUC 87.47% versus EfficientNet-trained-on-Google at 73.97%. In contrast, Chen et al. found the best machine learning model achieved AUC 0.985 versus human AUC 0.870, and Salini and HariKiran found approximately 70% of participants failed to detect fake videos while algorithms achieved over 90% accuracy.
- Interventions: raising awareness about deepfakes and financial incentives had no effect; warnings had mixed results; written detection tips did not improve accuracy; media literacy training and interactive training with example videos and/or feedback improved detection; providing AI predictions improved human detection performance.
Methods
- PRISMA systematic review registered on OSF; five databases (IEEE, ProQuest, PubMed, Web of Science, and Scopus) were searched in December 2022, with an updated search in December 2023; no time limits were imposed; unpublished works and dissertations were included; articles in languages other than English were excluded.
- Inclusion criteria: original study, reported in English, examined people’s detection of deepfakes, examined the influence of an intervention, strategy, or variable on deepfake detection, and reported relevant data needed to evaluate detection accuracy. Exclusions included AI-only detection studies, general media literacy without specific deepfake detection tasks, lack of criteria to assess detection accuracy, and synthetic stimuli not mimicking an existing person or retouching only. Thirty records with 40 independent studies were included.
- Narrative synthesis was conducted because substantial variability in design and outcomes made meta-analysis unsuitable; JBI Critical Appraisal tools were used for quasi-experimental, analytical cross-sectional, and qualitative studies.
Limitations
- Stated in the paper: the review focused on deepfakes imitating an existing person, excluding face retouching and face or voice synthesis, which limits generalizability; rapid evolution of deepfake technology may limit applicability to newer techniques; there is little research on audio-based deepfakes; the English-language restriction limits global representation and may overlook cultural and societal factors; most included studies were cross-sectional or experimental, so longitudinal conclusions about intervention effectiveness or changing detection abilities cannot be drawn.
- Evident from the text: performance metrics varied widely across studies, making comparison of human detection abilities difficult; risk-of-bias assessment found no experimental studies had follow-up, only two had pre-post measurements, and six lacked a control group or control condition; cross-sectional reporting issues included only one study having clear inclusion criteria, only half clearly describing subjects and research settings, and two not clearly identifying or controlling potential confounding factors.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
55 Valid · 7 Not valid · 38 Not checked · 81 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“For instance, studies have shown that under some circumstances people can surpass certain computer models in the detection of deepfakes [2].”
What the sources say
- Supportedcites [2] [2] M. Groh, Z. Epstein, C. Firestone, and R. Picard, ' Deepfake Detection by Human Crowds, Machines, and MachineInformed Crowds, ' Proceedings of the National Academy of Sciences 119, no. 1 (2022): e2110013119, https://doi.org/ 10.1073/pnas.2110013119. - source
The Discussion reports that participants significantly outperformed the model on videos of well-known political leaders. The Implications section likewise says participants performed better on the political videos and an attention-check video, supporting the claim’s qualification that this occurs under some circumstances. (LLM, conf=0.94)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Almost all video/image stimuli used face swapping or lip-sync technique, except for three studies that used morphing techniques [6, 7, 18].”
What the sources say
- Supportedcites [6] [6] R. S. Kramer, M. O. Mireku, T. R. Flack, and K. L. Ritchie, ' Face Morphing Attacks: Investigating Detection With Humans and Computers, ' Cognitive Research: Principles and Implications 4, no. 1 (2019): https://doi.org/10.1186/s41235019-0181-4. - source
The citation identifies this paper as one of the studies using morphing techniques, rather than asking it to establish the count across all studies. The Method > Materials passage describes morph images created with JPsychomorph and Adobe Photoshop and used as experimental stimuli. (LLM, conf=0.92)
- Supportedcites [7] [7] R. Nichols, C. Rathgeb, P. Drozdowski, and C. Busch, ' Psychophysical Evaluation of Human Performance in Detecting Digital Face Image Manipulations, ' IEEE Access 10 (2022): 31359 -31376, https://doi.org/10.1109/ACCESS.2022.3160596. - source
The citation serves to identify a study that used morphing stimuli. Section E, “Stimuli” (p. 12), explicitly lists morphing among the face manipulations used in this study; it does not establish the citing sentence’s count of three studies, which is not attributable to this source alone. (LLM, conf=0.91)
- Supportedcites [18] [18] D. J. Robertson, A. Mungall, D. G. Watson, K. A. Wade, S. J. Nightingale, and S. Butler, ' Detecting Morphed Passport Photos: A Training and Individual Di ff erences Approach, ' Cognitive Research: Principles and Implications 3 (2018): 1 -11, https://doi.org/10.1186/s41235-018-0113-8. - source
The citation identifies this paper as an example of a study using morphing techniques, rather than as evidence for the citing authors’ count of three studies. The Methods (Stimuli and apparatus) explicitly describe face morphs used as image stimuli. (LLM, conf=0.93)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Existing studies and reviews on deepfake detection have predominantly concentrated on AI-based detection methods [1, 11].”
- Also stated in: § High quality · plain
“As mentioned previously, most studies on deepfake detection focused on the e ffi cacy of detection algorithms.”
What the sources say
- Supportedcites [1] [1] R. Tolosana, R. Vera-Rodriguez, J. Fierrez, A. Morales, and J. Ortega-Garcia, ' Deepfakes and Beyond: A Survey of Face Manipulation and Fake Detection, ' Information Fusion 64 (2020): 131 -148, https://doi.org/10.1016/j.in ff us.2020.06.014. - source
The survey’s Manipulation Detection section reviews numerous AI-based detectors, including RNNs, SVMs and CNNs, and states that most attribute-manipulation detection systems are based on deep learning. This supports the broad characterization of the field’s research emphasis, although the paper does not quantify the share of reviews with that focus. Page numbers are unavailable in our copy. (LLM, conf=0.85)
- Not checked — no full text of the sourcecites [11] [11] L. A. Passos, D. Jodas, K. A. da Costa, L. A. S. Júnior, D. Colombo, and J. P. Papa, ' A Review of Deep LearningBased Approaches for Deepfake Content Detection ' 2022, https://arxiv.org/abs/2202.06095. - source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § High quality · plain
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Ethnicity has no e ff ect on detection ability [42]”
- Also stated in: § 3.3.1. Person-Level Factors · plain
“Multiple studies found that ethnicity does not impact deepfake detection.”
What the sources say
- Partly supportedcites [42] [42] S. Ahmed and H. W. Chua, ' Perception and Deception: Exploring Individual Responses to Deepfakes Across Di ff erent Modalities, ' Heliyon 9, no. 10 (2023)https://www.cell.com/ heliyon/fulltext/S2405-8440(23)07591-6. - sourceExperimental — accuracy not yet measured
Table 1 in the Results shows no statistically significant association between the study's White-versus-non-White race variable and perceived claim accuracy. But the study measured belief in a deepfake's claim, not deepfake detection ability, and a nonsignificant result does not establish that ethnicity has no effect. (LLM, conf=0.85)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 3.3.1. Person-Level Factors · plain
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Cognitive ability and analytical thinking are positively associated with detection accuracy [32, 23].”
- Also stated in: § 3.3.1. Person-Level Factors · hedged
“Cognitive ability and analytical thinking seem to be associated with better detection of political and celebrity deepfakes.”
What the sources say
- Not checked — no full text of the sourcecites [32] [32] S. Ahmed, ' Fooled by the Fakes: Cognitive Di ff erences in Perceived Claim Accuracy and Sharing Intention of Non-Political Deepfakes, ' Personality and Individual Differences 182 (2021): 111074, https://doi.org/10.1016/j.paid.2021.111074. - source
- Partly supportedcites [23] [23] M. Appel and F. Prietzel, ' The Detection of Political Deepfakes, ' Journal of Computer-Mediated Communication 27, no. 4 (2022): 13, https://doi.org/10.1093/jcmc/zmac008. - sourceExperimental — accuracy not yet measured
The General discussion says that participants who regularly engaged in analytic thinking were more likely to detect the deepfake in the first two studies. The Analytic thinking section describes the Cognitive Reflection Test as a measure of analytic thinking, but the cited passages do not establish a separate association between cognitive ability and detection accuracy. (LLM, conf=0.86)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 3.3.1. Person-Level Factors · hedged
82 more claims · 33 with their cited sources checked
Create a free account to see all 181 claims and to verify the 19 references still open.
- Probably not a substantial claim. 64
- It may state two separate claims as one. 48
- Borderline: one of the tests for a substantial claim is unclear. 1
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
“Specifically, how we judge a video's authenticity is not shaped solely by the sensory information it provides, but also by who we are as individuals, our prior knowledge, worldviews, cognitive styles, and even habitual media use (Somoray et al., 2025).”
From Deepfakes, Democracy, and the Ethics of Synthetic Media: A Synthesis of the SOLARIS Project (2026) · cites this paper as Somoray et al., 2025
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ High quality
“• The best algorithm model: 0.7397 vs. participants 0.08747 [36]”
§ High quality
“participants outperformed state-of-the-art machine learning detection models, achieving average AUC scores of 87.47% compared to E ffi cientNet-trained-on-Google at AUC of 73.97%.”
Citations
Cites
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021) · BMJ · doi:10.1136/bmj.n71 · Cited by 108,283
Lecture Notes in Computer Science 1205
(1999) · Industrial Robot: An International Journal · doi:10.1108/ir.1999.04926fae.001 · Cited by 38,662
JBI Manual for Evidence Synthesis
(2020) · JBI eBooks · doi:10.46658/jbimes-20-01 · Cited by 4,165
On the practice of dichotomization of quantitative variables.
MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D. (2002) · Psychological Methods · doi:10.1037/1082-989x.7.1.19 · Cited by 3,538
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
Lockwood, C., Munn, Z., & Porritt, K. (2015) · International Journal of Evidence-Based Healthcare · doi:10.1097/xeb.0000000000000062 · Cited by 2,456
Chapter 7: Systematic reviews of etiology and risk
Moola, S., Munn, Z., Tufanaru, C., Aromataris, E., Sears, K., Sfetic, R., Currie, M., Lisy, K., Qureshi, R., Mattis, P., et al. (2019) · JBI Reviewer's Manual · doi:10.46658/jbirm-17-06 · Cited by 2,114
Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems
(2020) · doi:10.1145/3313831 · Cited by 1,951
Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics
Li, Y., Yang, X., Sun, P., Qi, H., & Lyu, S. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00327 · Cited by 1,846
Chapter 7: Systematic reviews of etiology and risk
(2020) · JBI Manual for Evidence Synthesis · doi:10.46658/jbimes-20-08 · Cited by 1,268
Deepfakes and beyond: A Survey of face manipulation and fake detection
Tolosana, R., Vera-Rodriguez, R., Fierrez, J., Morales, A., & Ortega-Garcia, J. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.06.014 · Cited by 1,237
Vaccari, C., & Chadwick, A. (2020) · Social Media + Society · doi:10.1177/2056305120903408 · Cited by 978
Barker, T. H., Habibi, N., Aromataris, E., Stone, J. C., Leonardi-Bee, J., Sears, K., Hasanoff, S., Klugar, M., Tufanaru, C., Moola, S., et al. (2024) · JBI Evidence Synthesis · doi:10.11124/jbies-23-00268 · Cited by 843
A full systematic review was completed in 2 weeks using automation tools: a case study
Clark, J., Glasziou, P., Del Mar, C., Bannach-Brown, A., Stehlik, P., & Scott, A. M. (2020) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2020.01.008 · Cited by 687
Video Rewrite: driving visual speech with audio
Bregler, C., Covell, M., & Slaney, M. (1997) · Proceedings of the 24th annual conference on Computer graphics and interactive techniques - SIGGRAPH '97 · doi:10.1145/258734.258880 · Cited by 669
DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection
Jiang, L., Li, R., Wu, W., Qian, C., & Loy, C. C. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00296 · Cited by 599
Truth-Default Theory (TDT): A Theory of Human Deception and Deception Detection
Levine, T. R. (2014) · Journal of Language and Social Psychology · doi:10.1177/0261927x14535916 · Cited by 585
Masood, M., Nawaz, M., Malik, K. M., Javed, A., Irtaza, A., & Malik, H. (2023) · Applied Intelligence · doi:10.1007/s10489-022-03766-z · Cited by 543
Pragmatism vs interpretivism in qualitative information systems research
Goldkuhl, G. (2012) · European Journal of Information Systems · doi:10.1057/ejis.2011.54 · Cited by 541
FakeCatcher: Detection of Synthetic Portrait Videos using Biological Signals
Ciftci, U. A., Demir, I., & Yin, L. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.3009287 · Cited by 499
The deepfake detection challenge (dfdc) dataset
Dolhansky, B., Bitton, J., Pflaum, B., Lu, J., Howes, R., Wang, M., & Ferrer, C. C. (2020) · arXiv · doi:10.48550/arxiv.2006.07397 · Cited by 427
The Epistemic Threat of Deepfakes
Fallis, D. (2021) · Philosophy & Technology · doi:10.1007/s13347-020-00419-2 · Cited by 266
Do (microtargeted) deepfakes have real effects on political attitudes?
Dobber, T., Metoui, N., Trilling, D., Helberger, N., & de Vreese, C. (2021) · The International Journal of Press/Politics · doi:10.1177/1940161220944364 · Cited by 256
Deepfake detection by human crowds, machines, and machine-informed crowds
Groh, M., Epstein, Z., Firestone, C., & Picard, R. (2022) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2110013119 · Cited by 248
Fooled twice – people cannot detect deepfakes but think they can
Köbis, N., Doležalová, B., & Soraperra, I. (2021) · doi:10.2139/ssrn.3832978 · Cited by 194
Fooled twice: People cannot detect deepfakes but think they can
Köbis, N. C., Doležalová, B., & Soraperra, I. (2021) · iScience · doi:10.1016/j.isci.2021.103364 · Cited by 193
Content-Based Fake News Detection With Machine and Deep Learning: a Systematic Review
Capuano, N., Fenza, G., Loia, V., & Nota, F. D. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.02.005 · Cited by 183
A Review of Deep Learning-based Approaches for Deepfake Content Detection
Passos, L. A., Jodas, D., Costa, K. A. P., Júnior, L. A. S., Rodrigues, D., Ser, J. D., Camacho*, D., & Papa, J. (2023) · doi:10.22541/au.169735672.27713914/v1 · Cited by 116
A review of deep learning‐based approaches for deepfake content detection
(2024) · doi:10.1111/exsy.13570 · Cited by 116
Celeb-DF: A Large-scale Challenging Dataset for DeepFake ForensicsPreprint
Li, Y., Yang, X., Sun, P., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.12962 · Cited by 104
The detection of political deepfakes
Appel, M., & Prietzel, F. (2022) · Journal of Computer-Mediated Communication · doi:10.1093/jcmc/zmac008 · Cited by 95
Deepfakes and scientific knowledge dissemination
Doss, C., Mondschein, J., Shu, D., Wolfson, T., Kopecky, D., Fitton-Kane, V. A., Bush, L., & Tucker, C. (2023) · Scientific Reports · doi:10.1038/s41598-023-39944-3 · Cited by 90
Seeing is Believing: Exploring Perceptual Differences in DeepFake Videos
Tahir, R., Batool, B., Jamshed, H., Jameel, M., Anwar, M., Ahmed, F., Zaffar, M. A., & Zaffar, M. F. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445699 · Cited by 79
Ahmed, S. (2021) · Personality and Individual Differences · doi:10.1016/j.paid.2021.111074 · Cited by 71
Video Rewrite: Driving Visual Speech with Audio
Bregler, C., Covell, M., & Slaney, M. (2023) · Seminal Graphics Papers: Pushing the Boundaries, Volume 2 · doi:10.1145/3596711.3596787 · Cited by 67
DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery DetectionPreprint
Jiang, L., Li, R., Wu, W., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.03024 · Cited by 65
Detecting morphed passport photos: a training and individual differences approach
Robertson, D. J., Mungall, A., Watson, D. G., Wade, K. A., Nightingale, S. J., & Butler, S. (2018) · Cognitive Research: Principles and Implications · doi:10.1186/s41235-018-0113-8 · Cited by 49
Using deepfakes for experiments in the social sciences - A pilot study
Eberl, A., Kühn, J., & Wolbring, T. (2022) · Frontiers in Sociology · doi:10.3389/fsoc.2022.907199 · Cited by 48
Face morphing attacks: Investigating detection with humans and computers
Kramer, R. S. S., Mireku, M. O., Flack, T. R., & Ritchie, K. L. (2019) · Cognitive Research: Principles and Implications · doi:10.1186/s41235-019-0181-4 · Cited by 48
Providing detection strategies to improve human detection of deepfakes: An experimental study
Somoray, K., & Miller, D. J. (2023) · Computers in Human Behavior · doi:10.1016/j.chb.2023.107917 · Cited by 47
Self-citationThey Would Never Say Anything Like This! Reasons To Doubt Political Deepfakes
Hameleers, M., van der Meer, T. G. L. A., & Dobber, T. (2024) · European Journal of Communication · doi:10.1177/02673231231184703 · Cited by 47
Deepfake detection: humans vs. machines
Korshunov, P., & Marcel, S. (2020) · arXiv · doi:10.48550/arxiv.2009.03155 · Cited by 47
Seminal Graphics Papers: Pushing the Boundaries, Volume 2
(2023) · doi:10.1145/3596711 · Cited by 40
Ternovski, J., Kalla, J., & Aronow, P. M. (2021) · doi:10.31219/osf.io/dta97 · Cited by 37
(2020) · Communications in Computer and Information Science · doi:10.1007/978-3-030-63426-1 · Cited by 36
Perception and deception: Exploring individual responses to deepfakes across different modalities
Ahmed, S., & Chua, H. W. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e20383 · Cited by 29
Towards Understanding Perceptual Differences between Genuine and Face-Swapped Videos
Wöhler, L., Zembaty, M., Castillo, S., & Magnor, M. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445627 · Cited by 24
Ng, Y.-L. (2023) · Current Psychology · doi:10.1007/s12144-022-03621-x · Cited by 19
Psychophysical Evaluation of Human Performance in Detecting Digital Face Image Manipulations
Nichols, R., Rathgeb, C., Drozdowski, P., & Busch, C. (2022) · IEEE Access · doi:10.1109/access.2022.3160596 · Cited by 16
Sütterlin, S., Ask, T. F., Mägerle, S., Glöckler, S., Wolf, L., Schray, J., Chandi, A., Bursac, T., Khodabakhsh, A., Knox, B. J., et al. (2023) · Lecture Notes in Computer Science · doi:10.1007/978-3-031-35017-7_18 · Cited by 16
A Review of Deep Learning-based Approaches for Deepfake Content DetectionPreprint
Júnior, L. A. P., Jodas, D., da Costa, K. A. P., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2202.06095 · Cited by 15
Is it real? A study on detecting deepfake videos
Thaw, N. N., July, T., Wai, A. N., Goh, D. H., & Chua, A. Y. K. (2020) · Proceedings of the Association for Information Science and Technology · doi:10.1002/pra2.366 · Cited by 13
DeepFake Videos Detection Using Crowd Computing
Salini, Y., & HariKiran, J. (2024) · International Journal of Information Technology · doi:10.1007/s41870-023-01494-2 · Cited by 13
Exploring Neurophysiological Responses to Cross-Cultural Deepfake Videos
Khan, M. R., Naeem, S., Tariq, U., Dhall, A., Khan, M. N. A., Al Shargie, F., & Al-Nashash, H. (2023) · International Cconference on Multimodal Interaction · doi:10.1145/3610661.3617148 · Cited by 10
Trusted Media Challenge Dataset and User Study
Chen, W., Chua, S. L. B., Winkler, S., & Ng, S.-K. (2022) · Proceedings of the 31st ACM International Conference on Information & Knowledge Management · doi:10.1145/3511808.3557715 · Cited by 7
The Effect of Media Literacy on Misinformation and Deep Fake Video Detection
Shawky El Mokadem, S. (2023) · Arab Media & Society · doi:10.70090/sm23emlm · Cited by 6
Can Deepfakes be created on a whim?
Mehta, P., Jagatap, G., Gallagher, K., Timmerman, B., Deb, P., Garg, S., Greenstadt, R., & Dolan-Gavitt, B. (2023) · Companion Proceedings of the ACM Web Conference 2023 · doi:10.1145/3543873.3587581 · Cited by 5
(2022) · doi:10.1007/978-3-031-08147-7_24 · Cited by 5
Foundations and Practice of Security
Aı̈meur, E., Laurent, M., Yaich, R., et al. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-08147-7 · Cited by 4
Son, K. (2022) · University of Michigan · doi:10.7302/6457
Research works printed but not linked: 2
- Wöhler L., 2020, Computer animation and social agents, 120
- Ciftci U. A., 2020, Proceedings of the IEEE Transactions on Pattern Analysis and Machine Intelligence
Cited by
Deepfake detection: critical review of state-of-the-art approaches and future perspectives
Soundarya, B. C., & Gururaj, H. L. (2026) · Discover Applied Sciences · doi:10.1007/s42452-025-08174-9 · Cited by 25
The harm of deepfakes: a scoping review of deepfakes’ negative effects on human mind and behavior
Diel, A., Lalgi, T. J., Mellis, F. S., Teufel, A., & Bäuerle, A. (2025) · AI & Society · doi:10.1007/s00146-025-02774-0 · Cited by 22
Ismail, F. S. M., Diwan, T., & Pikle, N. K. (2026) · Cybernetics & Systems · doi:10.1080/01969722.2026.2654013 · Cited by 1
Seeing, Hearing, and Knowing Together: Multimodal Strategies in Deepfake Videos Detection
Chen, C., & Goh, D. H. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3790480 · Cited by 2
Oh, H., Lee, E., Shin, S., Lee, K., Pyun, M., & Lim, H. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791492 · Cited by 4
Metadata
- Authors
- Somoray, K., Miller, D. J., & Holmes, M.
- Affiliations
- James Cook University (AU)
- Corresponding author
- Somoray, Klaire
- Venue
- Human Behavior and Emerging Technologies
- Publisher
- Wiley
- Published
- 2025-01
- Volume
- 2025
- Issue
- 1
- ISSN
- 2578-1863 (ISSN-L)
- OpenAlex
- W4412881967
- Cited by
- 19
- References
- 41
- Citations by year
- 2026: 17 · 2025: 2
- FWCI
- 2.01
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0/ (vor, 214 d)
- doi.wiley.com/10.1002/tdm_license_1.1 (tdm)
- Locations
- Human Behavior and Emerging Technologies (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Psychology › Social Psychology › Deception detection and forensic psychology
- Keywords
- deepfake detection, synthetic media, human-AI comparison, detection accuracy, cognitive ability, analytical thinking, familiarity, human-AI collaboration
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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.