Wang, X.
Lomonosov Moscow State University
Vartanov, A.V.
Lomonosov Moscow State University
Li, Zh.
Lomonosov Moscow State University
Wang, Sh.
Lomonosov Moscow State University
Abstract
Background. In recent years, artificial intelligence (AI) technologies have rapidly permeated human society, and AI-generated faces have arguably become one of the most attention-grabbing technologies. Currently, Generative Adversarial Networks (GANs) have achieved unprecedented levels of realism in synthetic human faces. Behavioural data shows that humans can no longer reliably distinguish generated faces from real human faces — ushering in an era of AI and surrealism. However, the fact that there are no behavioural differences does not mean that the brain does not process these faces differently at the neural level. Consequently, applying the Event-Related Potential (ERP) method to study the bottom-up perceptual processing of faces and to investigate whether the brain can reflect the differences between AI-generated and real faces is an urgent research task.
Objectives.
Investigation of whether the brain differentiates between AI-generated and real human faces, using the ERP method and source localisation analysis.
Study Participants. The experiment involved 18 Chinese students studying at universities in Russia: 9 males and 9 females (mean age = 24 years, SD = 1.24). All participants were right-handed. Also, another 50 participants, Chinese students also studying at Russian universities, took part in a remote preliminary survey (23 men and 27 women; mean age = 24.54 years, SD = 2.23).
Methods. Participants were presented with 10 images of real Russian male celebrity faces and 10 AI-generated Slavic male faces. Each face was repeated 10 times, resulting in a total of 200 times. Electroencephalography (EEG) was recorded throughout. To process and analyse the collected EEG data, the brain activity source localisation method “Virtually Implanted Electrode” (Patent RU No. 2785268), developed by A.V. Vartanov, was applied.
Results. ERP and source localisation results revealed that the perception of AI-generated faces and real faces differed significantly in the right hemisphere around 180 ms. This difference was localised to areas VO1, VO2, V1, OFA, the hippocampus, and angular gyrus, and was highly correlated with the N170 component.
Conclusions. The brain is more sensitive to differences between AI-generated and real faces than is accessible to conscious perception.
References
Barabashnikov, V.A., Marinova, M.M. (2021). Deepfakes in research on facial perception. Ehksperimental’naya psikhologiya = Experimental Psychology, 14(1), 4–19. (In Russ.).
https://doi.org/10.17759/exppsy.2021000001
Bentin, S., Allison, T., Puce, A., Perez, E., McCarthy, G. (1996). Electrophysiological Studies of Face Perception in Humans. Journal of Cognitive Neuroscience, 8(6), 551–565.
https://doi.org/10.1162/jocn.1996.8.6.551
Brewer, A.A., Liu, J., Wade, A.R., Wandell, B.A. (2005). Visual field maps and stimulus selectivity in human ventral occipital cortex. Nature Neuroscience, 8(8), 1102–1109.
https://doi.org/10.1038/nn1507
Brooks, J.A., Freeman, J.B. (2026). Psychology and Neuroscience of Person Perception. In: J.T. Wixted, (ed.). Stevens’ Handbook of Experimental Psychology and Cognitive Neuroscience. (pp. 1–36). New Jersey: John Wiley & Sons Publ.
https://doi.org/10.1002/9781119170174.epcn413
Caporusso, N., Zhang, K., Carlson, G., Jachetta, D., Patchin, D. et al. (2020). User Discrimination of Content Produced by Generative Adversarial Networks. In: T. Ahram, R. Taiar, S. Colson, A. Choplin, (eds.). Human Interaction and Emerging Technologies. (pp. 725–730). Cham: Springer International Publishing.
https://doi.org/10.1007/978-3-030-25629-6_113
Dawel, A., Miller, E.J., Horsburgh, A., Ford, P. (2022). A systematic survey of face stimuli used in psychological research 2000–2020. Behavior Research Methods, 54(4), 1889–1901.
https://doi.org/10.3758/s13428-021-01705-3
Diel, A., MacDorman, K.F. (2021). Creepy cats and strange high houses: Support for configural processing in testing predictions of nine uncanny valley theories. Journal of Vision, 21(4), 1.
https://doi.org/10.1167/jov.21.4.1
Eiserbeck, A., Maier, M., Baum, J., Abdel Rahman, R. (2023). Deepfake smiles matter less-the psychological and neural impact of presumed AI-generated faces. Scientific Reports, 13(1), 16111.
https://doi.org/10.1038/s41598-023-42802-x
Freeman, J.B., Ambady, N. (2011). A dynamic interactive theory of person construal. Psychological Review, 118(2), 247–279.
https://doi.org/10.1037/a0022327
Freeman, J.B., Stolier, R.M., Brooks, J.A., Stillerman, B.S. (2018). The neural representational geometry of social perception. Current Opinion in Psychology, 24, 83–91.
https://doi.org/10.1016/j.copsyc.2018.10.003
Freeman, J.B., Stolier, R.M., Ingbretsen, Z.A., Hehman, E.A. (2014). Amygdala responsivity to high-level social information from unseen faces. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 34(32), 10573–10581.
https://doi.org/10.1523/JNEUROSCI.5063-13.2014
Goldfarb, E.V., Chun, M.M., Phelps, E.A. (2016). Memory-Guided Attention: Independent Contributions of the Hippocampus and Striatum. Neuron, 89(2), 317–324.
https://doi.org/10.1016/j.neuron.2015.12.014
Gonzalez-Franco, M., Bellido, A.I., Blom, K.J., Slater, M., Rodriguez-Fornells, A. (2016). The Neurological Traces of Look-Alike Avatars. Frontiers in Human Neuroscience, 10, 392.
https://doi.org/10.3389/fnhum.2016.00392
Halit, H., de Haan, M., Johnson, M.H. (2003). Cortical specialisation for face processing: Face-sensitive event-related potential components in 3- and 12-month-old infants. NeuroImage, 19(3), 1180–1193.
https://doi.org/10.1016/S1053-8119(03)00076-4
Haxby, J.V., Gobbini, M.I. (2011). 93 Distributed Neural Systems for Face Perception. In: A.J. Calder, G. Rhodes, M.H. Johnson, J.V. Haxby, (eds.). Oxford Handbook of Face Perception. Oxford: Oxford University Press.
https://doi.org/10.1093/oxfordhb/9780199559053.013.0006
Haxby, J.V., Hoffman, E.A., Gobbini, M.I. (2000). The distributed human neural system for face perception. Trends in Cognitive Sciences, 4(6), 223–233.
https://doi.org/10.1016/s1364-6613(00)01482-0
Kätsyri, J., de Gelder, B., de Borst, A.W. (2020). Amygdala responds to direct gaze in real but not in computer-generated faces. NeuroImage, 204, 116216.
https://doi.org/10.1016/j.neuroimage.2019.116216
Kegel, L.C., Brugger, P., Frühholz, S., Grunwald, T., Hilfiker, P. et al. (2020). Dynamic human and avatar facial expressions elicit differential brain responses. Social Cognitive and Affective Neuroscience, 15(3), 303–317.
https://doi.org/10.1093/scan/nsaa039
Kryukov, V.I. (2005). A model of attention and memory based on the principle of the dominant and the comparator function of the hippocampus. Neuroscience and Behavioral Physiology, 35(3), 235–252.
Miller, E.J., Foo, Y.Z., Mewton, P., Dawel, A. (2023a). How do people respond to computer-generated versus human faces? A systematic review and meta-analyses. Computers in Human Behavior Reports, 10, 100283.
https://doi.org/10.1016/j.chbr.2023.100283
Miller, E.J., Steward, B.A., Witkower, Z., Sutherland, C.A.M., Krumhuber, E.G., Dawel, A. (2023b). AI Hyperrealism: Why AI Faces Are Perceived as More Real Than Human Ones. Psychological Science, 34(12), 1390–1403.
https://doi.org/10.1177/09567976231207095
Momen, A., Hugenberg, K., Wiese, E. (2024). Social perception of robots is shaped by beliefs about their minds. Scientific Reports, 14(1), 5459.
https://doi.org/10.1038/s41598-024-53187-w
Mori, M., MacDorman, K.F., Kageki, N. (2012). The uncanny valley. IEEE Robotics & Automation Magazine, 19, 98–100.
https://doi.org/10.1109/MRA.2012.2192811
Moser, E., Derntl, B., Robinson, S., Fink, B., Gur, R.C., Grammer, K. (2007). Amygdala activation at 3T in response to human and avatar facial expressions of emotions. Journal of Neuroscience Methods, 161(1), 126–133.
https://doi.org/10.1016/j.jneumeth.2006.10.016
Mühlberger, A., Wieser, M.J., Herrmann, M.J., Weyers, P., Tröger, C., Pauli, P. (2009). Early cortical processing of natural and artificial emotional faces differs between lower and higher socially anxious persons. Journal of Neural Transmission, 116(6), 735–746.
https://doi.org/10.1007/s00702-008-0108-6
Mundy, M.E., Downing, P.E., Graham, K.S. (2012). Extrastriate cortex and medial temporal lobe regions respond differentially to visual feature overlap within preferred stimulus category. Neuropsychologia, 50(13), 3053–3061.
https://doi.org/10.1016/j.neuropsychologia.2012.07.006
Muukkonen, I., Ölander, K., Numminen, J., Salmela, V.R. (2020). Spatio-temporal dynamics of face perception. NeuroImage, 209, 116531.
https://doi.org/10.1016/j.neuroimage.2020.116531
Nightingale, S.J., Farid, H. (2022). AI-synthesized faces are indistinguishable from real faces and more trustworthy. Proceedings of the National Academy of Sciences, 119(8), e2120481119.
https://doi.org/10.1073/pnas.2120481119
Olivares, E.I., Iglesias, J., Saavedra, C., Trujillo-Barreto, N.J., Valdés-Sosa, M. (2015). Brain Signals of Face Processing as Revealed by Event-Related Potentials. Behavioural Neurology, 2015, 514361.
https://doi.org/10.1155/2015/514361
Pataranutaporn, P., Liu, R., Finn, E., Maes, P. (2023). Influencing human — AI interaction by priming beliefs about AI can increase perceived trustworthiness, empathy and effectiveness. Nature Machine Intelligence, 5(10), 1076–1086.
https://doi.org/10.1038/s42256-023-00720-7
Riedl, R., Mohr, P., Kenning, P., Davis, F., Heekeren, H. (2014). (PDF) Trusting Humans and Avatars: A Brain Imaging Study Based on Evolution Theory. Journal of Management Information Systems, 30(4), 83–114.
https://doi.org/10.2753/MIS0742-1222300404
Rossion, B. (2014). Understanding face perception by means of human electrophysiology. Trends in Cognitive Sciences, 18(6), 310–318.
https://doi.org/10.1016/j.tics.2014.02.013
Rossion, B., Jacques, C., Jonas, J. (2018). Mapping face categorization in the human ventral occipitotemporal cortex with direct neural intracranial recordings. Annals of the New York Academy of Sciences, (4), 1426.
https://doi.org/10.1111/nyas.13596
Sapronov, F.A., Gorbunova, E.S. (2025). A comparison of AI-generated stimuli and photographs: a study of visual search. Lomonosov Psychology Journal, 48(2), 109–131. (In Russ.).
https://doi.org/10.11621/LPJ-25
Schindler, S., Zell, E., Botsch, M., Kissler, J. (2017). Differential effects of face-realism and emotion on event-related brain potentials and their implications for the uncanny valley theory. Scientific reports, 7, 1–13.
https://doi.org/10.1038/srep45003
Schweinberger, S.R., Neumann, M.F. (2016). Repetition effects in human ERPs to faces. Cortex: a Journal Devoted to the Study of the Nervous System and Behavior, 80, 141–153.
https://doi.org/10.1016/j.cortex.2015.11.001
Shen, B., RichardWebster, B., O’Toole, A., Bowyer, K., Scheirer, W.J. (2021). A Study of the Human Perception of Synthetic Faces. arXiv, 2111.04230.
https://doi.org/10.48550/arXiv.2111.04230
Shigihara, Y., Zeki, S. (2014). Parallel processing of face and house stimuli by V1 and specialized visual areas: A magnetoencephalographic (MEG) study. Frontiers in Human Neuroscience, 8, 901.
https://doi.org/10.3389/fnhum.2014.00901
Studer, B., Cen, D., Walsh, V. (2014). The angular gyrus and visuospatial attention in decision-making under risk. NeuroImage, 103, 75–80.
https://doi.org/10.1016/j.neuroimage.2014.09.003
Varano, E., Vougioukas, K., Ma, P., Petridis, S., Pantic, M., Reichenbach, T. (2022). Speech-Driven Facial Animations Improve Speech-in-Noise Comprehension of Humans. Frontiers in Neuroscience, 15.
https://doi.org/10.3389/fnins.2021.781196
Vartanov, A.V. (2022). A new method of localizing brain activity using the scalp EEG data. Procedia Computer Science, 213, 41–48.
https://doi.org/10.1016/j.procs.2022.11.036
Vartanov, A.V. (2023). A new approach to the spatial localisation of electrical activity based on EEG data. Ehpilepsiya i paroksizmal’nye sostoyaniya = Epilepsy and Paroxysmal Conditions, 15(4), 326–338. (In Russ.).
https://doi.org/10.17749/2077-8333/epi.par.con.2023.177
Vasilieva, I.V., Asafova, A.N., Grischenko, D.S. (2025). Communicative personality characteristics and evaluation of AI-generated photos. Teoreticheskaya i ehksperimental'naya psikhologiya = Theoretical and Experimental Psychology, 18(1), 27–44. (In Russ.).
https://doi.org/10.11621/TEP-25-02
Vzorin, G.D., Bukinich, A.M., Sedykh, A.V, Vetrova, I.I., Sergienko, E.A. (2024). The Emotional Intelligence of the GPT-4 Large Language Model. Psychology in Russia: State of the Art, 17(2), 85–99.
https://doi.org/10.11621/pir.2024.0206
Wandell, B.A., Dumoulin, S.O., Brewer, A.A. (2007). Visual field maps in human cortex. Neuron, 56(2), 366–383.
https://doi.org/10.1016/j.neuron.2007.10.012
Wang, J., Chen, J., Yan, N., Wang, L., Ng, L. (2019). How do we perceive emotion from a 3-D virtual talking head: Evidence from functional near-infrared spectroscopy. International Conference on Interfaces and Human Computer Interaction, 115–122.
https://doi.org/10.33965/ihci2019_201906l015
Zhao, T., Chen, J., Wang, L., Yan, N. (2020). Emotional appraisal processing of computer-generated facial expressions: An functional near-infrared spectroscopy study. NeuroReport: For Rapid Communication of Neuroscience Research, 31(6), 437–441.
https://doi.org/10.1097/WNR.0000000000001420
To cite this article:
Wang, X., Vartanov, A.V. , Li, Zh., Wang, Sh..
Brain Mechanisms of Real Human and AI-Generated Face Perception. // National Psychological Journal
2026. 4. Pages145-157. doi: 10.11621/npj.2026.0410
Copied to Clipboard
Copy
