CoChat
Sign Up
Cited by 82Face Recognition and PerceptionFace recognition and analysisEvolutionary Psychology and Human Behavior

Deep models of superficial face judgments

Open access
Peterson, Joshua C., Uddenberg, Stefan, Griffiths, Thomas L., Todorov, Alexander, Suchow, Jordan W.·2022·Proceedings of the National Academy of Sciences
DOI 10.1073/pnas.2115228119PMC PMC9169911
Open this paper in CoChat — free account

Abstract

Significance: We quickly and irresistibly form impressions of what other people are like based solely on how their faces look. These impressions have real-life consequences ranging from hiring decisions to sentencing decisions. We model and visualize the perceptual bases of facial impressions in the most comprehensive fashion to date, producing photorealistic models of 34 perceived social and physical attributes (e.g., trustworthiness and age). These models leverage and demonstrate the utility of deep learning in face evaluation, allowing for 1) generation of an infinite number of faces that vary along these perceived attribute dimensions, 2) manipulation of any face photograph along these dimensions, and 3) prediction of the impressions any face image may evoke in the general (mostly White, North American) population.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1073/pnas.2115228119

Cites

5 of 5 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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 choice of using intelligence, trustworthiness, sociability and happiness as dependent attributes for the halo effect was driven by existing literature.

    “The choice of using intelligence, trustworthiness, sociability and happiness as dependent attributes for the halo effect was driven by existing literature on this cognitive bias, such as [4,6,8,9,22,53,60–63].”

    From What is beautiful is still good: the attractiveness halo effect in the era of beauty filters (2024) · cites this paper as [62]

    In Results, “The Structure of Attribute Inferences” (page unavailable), the paper analyzes ratings of smartness, trustworthiness, happiness, and outgoingness, which is close to sociability, and compares some findings with prior work. It does not describe these four attributes as dependent measures of a halo effect or say that literature on that bias motivated their selection. (LLM, conf=0.78)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 37

  1. [1]F. Farzin, C. Hou, A. M. Norcia, Piecing it together: Infants’ neural responses to face and object structure. J. Vis. 12, 6– 6 (2012). PMC3587016 23220577
  2. [2]N. Kanwisher, J. McDermott, M. M. Chun, The fusiform face area: A module in human extrastriate cortex specialized for face perception. J. Neurosci. 17, 4302– 4311 (1997). 9151747 PMC6573547
  3. [3]C. Frith, Role of facial expressions in social interactions. Philos. Trans. R. Soc. Lond. B Biol. Sci. 364, 3453– 3458 (2009). 19884140 PMC2781887
  4. [4]N. N. Oosterhof, A. Todorov, The functional basis of face evaluation. Proc. Natl. Acad. Sci. U.S.A. 105, 11087– 11092 (2008). 18685089 PMC2516255
  5. [5]C. A. Sutherland., Social inferences from faces: Ambient images generate a three-dimensional model. Cognition 127, 105– 118 (2013). 23376296
  6. [6]L. A. Zebrowitz, First impressions from faces. Curr. Dir. Psychol. Sci. 26, 237– 242 (2017). 28630532 PMC5473630
  7. [7]B. C. Jones., To which world regions does the valence-dominance model of social perception apply? Nat. Hum. Behav. 5, 159– 169 (2021). 33398150
  8. [9]C. A. M. Sutherland., Facial first impressions across culture: Data-driven modeling of Chinese and British perceivers’ unconstrained facial impressions. Pers. Soc. Psychol. Bull. 44, 521– 537 (2018). 29226785
  9. [10]A. Todorov, C. Y. Olivola, R. Dotsch, P. Mende-Siedlecki, Social attributions from faces: Determinants, consequences, accuracy, and functional significance. Annu. Rev. Psychol. 66, 519– 545 (2015). 25196277
  10. [11]A. Todorov, A. N. Mandisodza, A. Goren, C. C. Hall, Inferences of competence from faces predict election outcomes. Science 308, 1623– 1626 (2005). 15947187
  11. [13]I. V. Blair, C. M. Judd, K. M. Chapleau, The influence of Afrocentric facial features in criminal sentencing. Psychol. Sci. 15, 674– 679 (2004). 15447638
  12. [14]J. L. Eberhardt, P. G. Davies, V. J. Purdie-Vaughns, S. L. Johnson, Looking deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing outcomes. Psychol. Sci. 17, 383– 386 (2006). 16683924
  13. [15]C. J. Bohil, H. M. Kleider-Offutt, C. Killingsworth, A. M. Meacham, Training away face-type bias: Perception and decisions about emotional expression in stereotypically black faces. Psychol. Res. 85, 2727– 2741 (2021). 33074362
  14. [16]M. Turk, A. Pentland, Eigenfaces for recognition. J. Cogn. Neurosci. 3, 71– 86 (1991). 23964806
  15. [17]B. Tiddeman, M. Stirrat, D. Perrett, Towards realism in facial prototyping: Results of a wavelet mrf method. Proc. Theory Pract Comp. Graph. 1, 20– 30 (2006).
  16. [18]V. Blanz, T. Vetter, “A morphable model for the synthesis of 3d faces” in Proceedings of the 26th Annual Conference on Computer Graphics and Interactive Techniques, A. Rockwood, Ed. (ACM Press/Addison-Wesley Publishing Co., 1999), pp. 187–194.
  17. [20]O. M. Parkhi, A. Vedaldi, A. Zisserman, “ Deep face recognition” in Proceedings of the British Machine Vision Conference (BMVC), X. Xi, M. W. Jones, G. K. L. Tam, Eds. (BMVA Press, 2015), pp. 41.1– 41.12.
  18. [22]T. Karras., “Analyzing and improving the image quality of StyleGAN” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (IEEE, 2020), pp. 8110–8119.
  19. [24]A. J. O’Toole, C. D. Castillo, C. J. Parde, M. Q. Hill, R. Chellappa, Face space representations in deep convolutional neural networks. Trends Cogn. Sci. 22, 794– 809 (2018). 30097304
  20. [26]P. Terhörst, D. Fährmann, N. Damer, F. Kirchbuchner, A. Kuijper, “Beyond identity: What information is stored in biometric face templates?” in 2020 IEEE International Joint Conference on Biometrics (IJCB) (IEEE, 2020), pp. 1–10.
  21. [27]V. Bruce, A. Young, Understanding face recognition. Br. J. Psychol. 77, 305– 327 (1986). 3756376
  22. [28]D. Oh, R. Dotsch, J. Porter, A. Todorov, Gender biases in impressions from faces: Empirical studies and computational models. J. Exp. Psychol. Gen. 149, 323– 342 (2020). 31294585
  23. [29]J. R. Collova, C. A. M. Sutherland, G. Rhodes, Testing the functional basis of first impressions: Dimensions for children’s faces are not the same as for adults’ faces. J. Pers. Soc. Psychol. 117, 900– 924 (2019). 31169388
  24. [30]W. A. Bainbridge, P. Isola, A. Oliva, The intrinsic memorability of face photographs. J. Exp. Psychol. Gen. 142, 1323– 1334 (2013). 24246059
  25. [31]G. Rhodes, The evolutionary psychology of facial beauty. Annu. Rev. Psychol. 57, 199– 226 (2006). 16318594
  26. [32]C. P. Said, A. Todorov, A statistical model of facial attractiveness. Psychol. Sci. 22, 1183– 1190 (2011). 21852448
  27. [33]I. J. Goodfellow., “ Generative adversarial networks.” in Advances in Neural Information Processing Systems, Z. Ghahramani, M. Welling, C. Cortes, N. Lawrence, K. Q. Weinberger, Eds. (Curran Associates, Inc., 2014), vol. 27.
  28. [34]J. E. Martinez, F. Funk, A. Todorov, Quantifying idiosyncratic and shared contributions to judgment. Behav. Res. Methods 52, 1428– 1444 (2020). 31898288
  29. [36]C. J. Parde, Y. Hu, C. Castillo, S. Sankaranarayanan, A. J. O’Toole, Social trait information in deep convolutional neural networks trained for face identification. Cogn. Sci. 43, e12729 (2019). 31204800 PMC8735867
  30. [38]D. S. Ma, J. Correll, B. Wittenbrink, The Chicago face database: A free stimulus set of faces and norming data. Behav. Res. Methods 47, 1122– 1135 (2015). 25582810
  31. [39]A. Kiyonaga, J. M. Scimeca, D. P. Bliss, D. Whitney, Serial dependence across perception, attention, and memory. Trends Cogn. Sci. 21, 493– 497 (2017). 28549826 PMC5516910
  32. [41]D. Gragnaniello, D. Cozzolino, F. Marra, G. Poggi, L. Verdoliva, “Are GAN generated images easy to detect? A critical analysis of the state-of-the-art” in 2021 IEEE International Conference on Multimedia and Expo (ICME) (IEEE, 2021), pp. 1–6.
  33. [42]G. Tang., Detection of GAN-synthesized image based on discrete wavelet transform. Secur. Commun. Netw. 2021, 5511435 (2021).
  34. [43]R. Winick, Intellectual property, defamation and the digital alteration of visual images. Columbia VLA J. Law Arts 21, 143 (1996).
  35. [44]L. B. A. Potter, Altered realities: The effect of digital imaging technology on libel and right of privacy. Hast. Comm. Ent. LJ 17, 495 (1994).
  36. [47]L. Litman, J. Robinson, T. Abberbock, TurkPrime.com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behav. Res. Methods 49, 433– 442 (2017). 27071389 PMC5405057
  37. [48]A. Dosovitskiy, T. Brox, “ Generating images with perceptual similarity metrics based on deep networks” in Advances in Neural Information Processing Systems, D. Lee, M. Sugiyama, U. Luxburg, I. Guyon, R. Garnett, Eds. (Curran Associates, Inc., 2016), pp. 658– 666.

Cited by

Sorted by citations

Authors
Peterson, J. C., Uddenberg, S., Griffiths, T. L., Todorov, A., & Suchow, J. W.
Affiliations
  • Princeton University (US)
  • University of Chicago (US)
  • Stevens Institute of Technology (US)
Corresponding author
Peterson, Joshua C.
ORCID
Venue
Proceedings of the National Academy of Sciences
Publisher
National Academy of Sciences
Published
2022-04-21
Volume
119
Issue
17
Pages
e2115228119-e2115228119
ISSN
  • 0027-8424 (ISSN-L)
  • 1091-6490
DOI
10.1073/pnas.2115228119
PMC
PMC9169911
OpenAlex
W4224222182
Cited by
82
References
5
Citations by year
2026: 13 · 2025: 14 · 2024: 23 · 2023: 22 · 2022: 10
FWCI
8.90
Type
journal article
Language
en
Open access
Licence
Locations
  • Proceedings of the National Academy of Sciences (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • Knowledge@UChicago (University of Chicago) (submittedVersion, cc-by-nc-nd)
  • University of Chicago
Funders
  • Princeton University (100006734)
  • UChicago | University of Chicago Booth School of Business (100007618)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Evolutionary Psychology and Human Behavior
Keywords
face perception, deep generative models, trait inference, trustworthiness, age estimation, adiposity, synthetic face generation, inter-rater reliability
MeSH terms
Attitude, Face, Facial Expression, Humans, Judgment, Social Perception, Trust

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