pith. sign in

arxiv: 2605.17461 · v1 · pith:Z3GWZ76Inew · submitted 2026-05-17 · 💻 cs.HC · cs.AI· cs.CY

Artificial Intelligence can Recognize Whether a Job Applicant is Selling and/or Lying According to Facial Expressions and Head Movements Much More Correctly Than Human Interviewers

Hung-Yue Suen , Kuo-En Hung , Che-Wei Liu , Yu-Sheng Su , Han-Chih Fan This is my paper

Pith reviewed 2026-05-19 22:44 UTC · model grok-4.3

classification 💻 cs.HC cs.AIcs.CY
keywords impression managementdeception detectionfacial expressionshead movementsdeep learningvideo interviewscomputer visionAI performance comparison
0
0 comments X p. Extension
pith:Z3GWZ76I Add to your LaTeX paper What is a Pith Number? 
\usepackage{pith}
\pithnumber{Z3GWZ76I}

Prints a linked pith:Z3GWZ76I badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

The pith

Deep learning models detect honest and deceptive impression management in job interviews from facial expressions and head movements more accurately than human interviewers.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper trains deep learning models on videos of 121 job applicants responding to structured behavioral questions in asynchronous interviews. The models extract temporal patterns in facial expressions and head movements to match the applicants' own post-interview self-reports on four measures of honest and deceptive impression management tactics. In a side-by-side test on 30 videos, the models produced stronger correlations with those self-reports than 30 human interviewers achieved. If correct, this would mean automated analysis can supply more consistent evaluations of how candidates present themselves than human judgment alone. Readers care because hiring decisions often hinge on assessing authenticity in ways that are hard for people to do reliably.

Core claim

Deep learning models enabled by computer vision extract temporal patterns of facial expressions and head movements from video frames in real asynchronous video interviews to identify self-reported honest and deceptive impression management tactics, explaining 91% of the variance in honest IM and 84% in deceptive IM while showing stronger correlations with self-reported IM scores than human interviewers.

What carries the argument

Deep learning models using computer vision to extract temporal patterns of facial expressions and head movements from video frames for predicting impression management scores.

If this is right

  • Automated video analysis can predict honest and deceptive impression management with higher concurrent validity than human evaluators.
  • Structured behavioral interview videos contain sufficient visual signals for models to account for most of the variance in self-reported IM tactics.
  • AI approaches offer a scalable alternative for screening applicant trustworthiness in asynchronous interview formats.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Widespread use could shift hiring toward standardized visual screening and reduce differences in how individual interviewers interpret the same cues.
  • The same modeling approach might be tested on synchronous live interviews or extended to other professional interactions like sales calls.
  • Organizations would need policies for consent and data handling when applying facial analysis to employment decisions.

Load-bearing premise

Self-reported impression management measures from post-interview surveys serve as accurate and unbiased ground truth for training the models and for comparing them to human interviewers.

What would settle it

An independent study that measures actual deceptive behaviors through methods such as reference checks or follow-up performance data and finds the models do not outperform humans in predicting those verified behaviors.

Figures

Figures reproduced from arXiv: 2605.17461 by Che-Wei Liu, Han-Chih Fan, Hung-Yue Suen, Kuo-En Hung, Yu-Sheng Su.

Figure 1
Figure 1. Figure 1: Face mesh topology with facial landmarks. The MediaPipe Face Mesh utilizes two interconnected NN models to deduce the 3D surface geometry of a human face from a single smartphone or webcam input, eliminating the need for specialized depth sensors. One model processes the entire image to identify face locations, while the other focuses on these located faces to predict their 3D surface geometry using regres… view at source ↗
Figure 2
Figure 2. Figure 2: Eye corner alignment To convert the original 2D images into a vector of 3D coordinates, we transformed these images into 3D space using a Cartesian coordinate system [54], as shown in [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: 2D to 3D image transformation We obtained P(𝑋𝑓𝑚, 𝑌𝑓𝑚, 𝑍𝑓𝑚) from FaceMesh, where the y￾coordinate matches the previously mentioned horizontal axis. The x and y coordinates of the vertices match the points in the 2D image. We rotated the 2D images along the x-, y-, and z￾axes using the Cartesian coordinate system. The height (h) and width (w) of the image were calculated as depicted in equation (1), while th… view at source ↗
Figure 4
Figure 4. Figure 4: Proposed Image Processing and 3D-CNN - LSTM Regression Models IV. RESULTS A. Self-reported IM: Statistics, Validity, and Reliability To assess the construct validity and reliability of our model for self-reported IMs — including honest self-promotion, honest [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
read the original abstract

Whether an interviewee's honest and deceptive responses can be detected by facial expression signals in videos has been debated and requires further research. We developed deep learning models enabled by computer vision to extract temporal patterns of job applicants' facial expressions and head movements to identify self-reported honest and deceptive impression management (IM) tactics from video frames in real asynchronous video interviews. A 12- to 15-minute video was recorded for each of N=121 job applicants as they answered five structured behavioral interview questions. Each applicant completed a survey to self-evaluate their trustworthiness on four IM measures. Additionally, a field experiment was conducted to compare the concurrent validity associated with self-reported IMs between our modeling approach and human interviewers. Human interviewers' performance in predicting these IM measures from another subset of 30 videos was obtained by having N=30 human interviewers evaluate three recordings. Our models explained 91% and 84% of the variance in honest and deceptive IMs, respectively, and showed stronger correlations with self-reported IM scores than human interviewers.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper claims that deep learning models using computer vision on facial expressions and head movements from 12-15 minute asynchronous video interviews (N=121 applicants answering five behavioral questions) can predict self-reported honest and deceptive impression management (IM) tactics with R²=0.91 and 0.84 respectively, and that these models show stronger correlations with the self-reported IM scores than N=30 human interviewers evaluating a subset of 30 videos.

Significance. If the self-reported IM scores validly index objective behavior, the high variance explained and outperformance of humans could support automated screening tools in hiring. The results highlight potential for temporal pattern analysis in video interviews, but significance is limited by the absence of independent validation that self-reports track actual lying or selling rather than self-perception.

major comments (3)
  1. [Abstract] Abstract: The title and abstract assert that the models recognize 'whether a job applicant is selling and/or lying' from facial/head signals, yet the models are trained and evaluated exclusively against post-interview self-reported IM scores with no independent behavioral coding, physiological measures, or third-party ratings to establish that these scores index actual deceptive tactics rather than social-desirability bias or inaccurate self-insight.
  2. [Field experiment] Field experiment section: The concurrent validity comparison shows models outperforming humans in predicting the same self-reported IM targets (N=30 videos), but this does not demonstrate superior detection of objective IM because both the AI and human judgments are benchmarked against the identical self-report criterion; superior correlation with self-reports does not entail better recognition of actual lying/selling.
  3. [Results] Results: The reported R² values of 91% and 84% on N=121 samples are presented without details on train-test splits, cross-validation procedure, or regularization to control overfitting in the deep learning models, which is load-bearing for interpreting the performance as generalizable rather than label-fitting on the self-report targets.
minor comments (2)
  1. [Abstract] Abstract and methods: No description of the specific computer vision pipeline, model architecture (e.g., CNN/RNN layers), or feature extraction for temporal facial and head movement patterns.
  2. [Discussion] The manuscript lacks discussion of prior literature on the known limitations of self-report IM measures in interview contexts or explicit acknowledgment that the ground truth is perceptual rather than behavioral.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for these insightful comments, which help clarify the boundaries of our claims. We address each major point below, acknowledging where revisions are needed to better align language with the self-report criterion used throughout the study.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The title and abstract assert that the models recognize 'whether a job applicant is selling and/or lying' from facial/head signals, yet the models are trained and evaluated exclusively against post-interview self-reported IM scores with no independent behavioral coding, physiological measures, or third-party ratings to establish that these scores index actual deceptive tactics rather than social-desirability bias or inaccurate self-insight.

    Authors: We agree that the title and abstract phrasing could be more precise. The models were developed and evaluated to predict self-reported honest and deceptive impression management scores, which serve as the criterion variable in line with prior organizational psychology research on IM tactics. We will revise the title and abstract to explicitly state that the models predict self-reported IM rather than claiming direct recognition of objective lying or selling behavior, and we will add a sentence in the discussion noting the reliance on self-reports as a limitation. revision: yes

  2. Referee: [Field experiment] Field experiment section: The concurrent validity comparison shows models outperforming humans in predicting the same self-reported IM targets (N=30 videos), but this does not demonstrate superior detection of objective IM because both the AI and human judgments are benchmarked against the identical self-report criterion; superior correlation with self-reports does not entail better recognition of actual lying/selling.

    Authors: We concur that the field experiment demonstrates superior concurrent validity with the self-reported IM scores rather than superior detection of objective behavior. The comparison is valuable for showing that the AI approach aligns more closely with applicants' own reports of their IM tactics than human interviewers do. We will revise the results and discussion sections to frame the finding explicitly as better prediction of self-reported IM and to avoid any implication of objective superiority without additional validation. revision: partial

  3. Referee: [Results] Results: The reported R² values of 91% and 84% on N=121 samples are presented without details on train-test splits, cross-validation procedure, or regularization to control overfitting in the deep learning models, which is load-bearing for interpreting the performance as generalizable rather than label-fitting on the self-report targets.

    Authors: This point is well taken. The current manuscript lacks sufficient detail on the modeling pipeline. In the revision we will add a dedicated subsection describing the train-test split strategy, cross-validation approach (including k-fold details), regularization methods, and any hyperparameter tuning used to address potential overfitting. We will also report performance on held-out test data to support the interpretability of the R² values. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is standard supervised learning on chosen labels

full rationale

The paper extracts facial and head-movement features from N=121 videos, trains models to predict the applicants' self-reported IM scores on four measures, reports R² values of 0.91/0.84, and shows higher correlation with those same self-report scores than human raters achieve on a separate set of 30 videos. This is a conventional supervised-learning pipeline whose performance metrics are defined directly against the chosen target variable. No equation reduces to its input by construction, no parameter is fitted on one subset and then presented as an independent prediction of a distinct quantity, and no self-citation or uniqueness theorem is invoked to justify the central mapping. The assumption that self-reports validly index objective lying/selling is a substantive validity claim rather than a definitional or statistical circularity; the reported derivation chain remains self-contained against the paper's own benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that facial and head movement patterns carry detectable signals of impression management, plus the untested premise that self-reports are faithful proxies; the deep learning component introduces numerous fitted parameters whose values are not reported.

free parameters (1)
  • deep learning model hyperparameters and weights
    Neural network parameters are optimized during training to maximize fit to the self-reported IM labels from the N=121 participants.
axioms (1)
  • domain assumption Facial expressions and head movements contain temporal patterns that reliably indicate honest and deceptive impression management tactics.
    Invoked to justify the computer vision extraction step as a valid input for identifying self-reported IM.

pith-pipeline@v0.9.0 · 5739 in / 1489 out tokens · 42968 ms · 2026-05-19T22:44:47.591123+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

73 extracted references · 73 canonical work pages · 1 internal anchor

  1. [1]

    Initial impressions: What they are, what they are not, and how they influence structured interview outcomes,

    B. W. Swider, M. R. Barrick, and T. B. Harris, "Initial impressions: What they are, what they are not, and how they influence structured interview outcomes," J. Appl. Psychol., vol. 101, no. 5, pp. 625–638, May 2016, doi: 10.1037/apl0000077

    work page doi:10.1037/apl0000077 2016

  2. [2]

    Why does impression management positively influence interview ratings? The mediating role of competence and warmth,

    A. A. Amaral, D. M. Powell, and J. L. Ho, "Why does impression management positively influence interview ratings? The mediating role of competence and warmth," Int. J. Selection Assessment, vol. 27, no. 4, pp. 315–327, Aug. 2019, doi: 10.1111/ijsa.12260

    work page doi:10.1111/ijsa.12260 2019

  3. [3]

    ‘Faking’ from the applicant's perspective: A theory of self - presentation in personnel selection settings,

    B. Marcus, "‘Faking’ from the applicant's perspective: A theory of self - presentation in personnel selection settings," Int. J. Selection Assessment, vol. 17 , no. 4, pp. 417 –430, Nov. 2009, doi: 10.1111/j.1468 - 2389.2009.00483.x

    work page doi:10.1111/j.1468 2009

  4. [4]

    “I (might be) just that good

    J. S. Bourdage, N. Roulin, and R. Tarraf, "“I (might be) just that good”: Honest and deceptive impression management in employment interviews," Pers. Psychol., vol. 71 , no. 4, pp. 597 –632, Aug. 2018, doi: 10.1111/peps.12285

    work page doi:10.1111/peps.12285 2018

  5. [5]

    Honest and deceptive impression management in the employment interview: Can it be detected and how does it impact evaluations?,

    N. Roulin, A. Bangerter, and J. Levashina, "Honest and deceptive impression management in the employment interview: Can it be detected and how does it impact evaluations?," Pers. Psychol., vol. 68, no. 2, pp. 395–444, Jun. 2015, doi: 10.1111/peps.12079

    work page doi:10.1111/peps.12079 2015

  6. [6]

    An updated survey of beliefs and practices related to faking in individual assessments,

    C. Robie, S. D. Risavy, R. R. Jacobs, N. D. Christiansen, C. J. Kö nig, and A. B. Speer, "An updated survey of beliefs and practices related to faking in individual assessments," Int. J. Selection Assessment, vol. 29, no. (3-4), pp. 503–509, May 2021, doi: 10.1111/ijsa.12335

    work page doi:10.1111/ijsa.12335 2021

  7. [7]

    LieToMe: An ensemble approach for deception detection from facial cues,

    D. Avola, M. Cascio, L. Cinque, A. Fagioli, and G. L. Foresti, "LieToMe: An ensemble approach for deception detection from facial cues," Int. J. Neural Syst., vol. 31 , no. 2, p. 2050068, Nov. 2021, doi: 10.1142/s0129065720500689

    work page doi:10.1142/s0129065720500689 2021

  8. [8]

    Pitfalls and opportunities in nonverbal and verbal lie detection,

    A. Vrij, P. A. Granhag, and S. Porter, "Pitfalls and opportunities in nonverbal and verbal lie detection," Psychological Sci. Public Interest, vol. 11, no. 3, pp. 89–121, Dec. 2010, doi: 10.1177/1529100610390861. [9]] D. B. Buller and J. K. Burgoon, "Interpersonal deception theory," Commun. Theory, vol. 6, no. 3, pp. 203 –242, Aug. 1996, doi: 10.1111/j.14...

    work page doi:10.1177/1529100610390861 2010

  9. [9]

    A video-based screening system for automated risk assessment using nuanced facial features,

    S. J. Pentland, N. W. Twyman, J. K. Burgoon, J. F. Nunamaker, and C. B. R. Diller, "A video-based screening system for automated risk assessment using nuanced facial features," J. Manag. Inf. Syst., vol. 34, no. 4, pp. 970– 993, Oct. 2017, doi: 10.1080/07421222.2017.1393304

    work page doi:10.1080/07421222.2017.1393304 2017

  10. [10]

    Accuracy of deception judgments,

    C. F. Bond and B. M. DePaulo, "Accuracy of deception judgments," Personality Social Psychol. Rev., vol. 10, no. 3, pp. 214 –234, Aug. 2006, doi: 10.1207/s15327957pspr1003_2

    work page doi:10.1207/s15327957pspr1003_2 2006

  11. [11]

    Identifying applicant faking in job interviews: Examining the role of criterion -based content analysis and storytelling,

    N. Roulin and D. M. Powell, "Identifying applicant faking in job interviews: Examining the role of criterion -based content analysis and storytelling," J. Pers. Psychol., vol. 17, no. 3, pp. 143–154, Jul. 2018, doi: 10.1027/1866-5888/a000207

    work page doi:10.1027/1866-5888/a000207 2018

  12. [12]

    Detecting deception from emotional and unemotional cues,

    G. Warren, E. Schertler, and P. Bull, "Detecting deception from emotional and unemotional cues," J. Nonverbal Behav., vol. 33 , no. 1, pp. 59 –69, Nov. 2009, doi: 10.1007/s10919-008-0057-7

    work page doi:10.1007/s10919-008-0057-7 2009

  13. [13]

    Spontaneous facial micro-expression recognition using 3D spatiotemporal convolutional neural networks,

    S. P. T. Reddy, S. T. Karri, S. R. Dubey, and S. Mukherjee, "Spontaneous facial micro-expression recognition using 3D spatiotemporal convolutional neural networks," in 2019 Int. Joint Conf. Neural Netw. (IJCNN), Budapest, Hungary, Jul. 14-19 2019, pp. 1–8

    work page 2019

  14. [14]

    Face-focused cross-stream network for deception detection in videos,

    M. Ding, A. Zhao, Z. Lu, T. Xiang, and J. Wen, "Face-focused cross-stream network for deception detection in videos," in 2019 IEEE/CVF Conf. Comput. Vision Pattern Recognit. (CVPR) , Long Beach, CA, Jun. 15 -20 2019, pp. 7794–7803

    work page 2019

  15. [15]

    Toward end -to-end deception detection in videos,

    H. Karimi, J. Tang, and Y. Li, "Toward end -to-end deception detection in videos," in 2018 IEEE Int. Conf. Big Data, Seattle, WA, Dec. 10-13 2018, pp. 1278–1283

    work page 2018

  16. [16]

    Deception detection in videos using robust facial features,

    A. Stathopoulos, L. Han, N. Dunbar, J. K. Burgoon, and D. Metaxas, "Deception detection in videos using robust facial features," in Proc. Future Technol. Conf. (FTC) 2020, Volume 3, Cham, Switzerland, Nov. 5- 6 2021, pp. 668–682

    work page 2020

  17. [17]

    Deception detection in videos,

    Z. Wu, B. Singh, L. Davis, and V. Subrahmanian, "Deception detection in videos," Proc. AAAI Conf. Artif. Intell., vol. 32, no. 1, pp. 1695-1702, Apr. 2018, doi: 10.1609/aaai.v32i1.11502

    work page doi:10.1609/aaai.v32i1.11502 2018

  18. [18]

    Into the void: A conceptual model and research agenda for the design and use of asynchronous video interviews,

    E. R. Lukacik, J. S. Bourdage, and N. Roulin, "Into the void: A conceptual model and research agenda for the design and use of asynchronous video interviews," Hum. Resour. Manag. Rev., vol. 32, no. 1, 100789, Mar. 2022, doi:10.1016/j.hrmr.2021.100789

    work page doi:10.1016/j.hrmr.2021.100789 2022

  19. [19]

    Revealing the Influence of AI and Its Interfaces on Job Candidates' Honest and Deceptive Impression Management in Asynchronous Video Interviews,

    H. -Y. Suen and K. -E. Hung, “ Revealing the Influence of AI and Its Interfaces on Job Candidates' Honest and Deceptive Impression Management in Asynchronous Video Interviews," Technol. Forecast. Soc. Change, vol. 198, no. 123011, Jan. 2024, doi: 10.1016/j.techfore.2023.123011

    work page doi:10.1016/j.techfore.2023.123011 2024

  20. [20]

    Comparing job applicant deception in asynchronous vs synchronous video interviews, with and without AI - assisted assessments,

    H. -Y. Suen and K. -E. Hung, "Comparing job applicant deception in asynchronous vs synchronous video interviews, with and without AI - assisted assessments," Inf. Technol. People, vol. ahead-of-print, no. ahead- of-print, 2023, doi:10.1108/ITP-02-2023-0189

    work page doi:10.1108/itp-02-2023-0189 2023

  21. [21]

    A Review of Dynamic Datasets for Facial Expression Research,

    E.G. Krumhuber, L. Skora, D. Kü ster, and L. Fou, "A Review of Dynamic Datasets for Facial Expression Research," Emotion Rev., vol. 9, no. 3, pp. 280-292, Oct. 2017, doi:10.1177/1754073916670022

    work page doi:10.1177/1754073916670022 2017

  22. [22]

    Technology in the Employment Interview: A Meta -Analysis and Future Research Agenda,

    N. Blacksmith, J. C. Willford, and T. S. Behrend, "Technology in the Employment Interview: A Meta -Analysis and Future Research Agenda," Personnel Assess. Decis ., vol. 1, no. 2, Jun. 2016, doi:10.25035/pad.2016.00

    work page doi:10.25035/pad.2016.00 2016

  23. [23]

    Development of an AI-Based Interview System for Remote Hiring,

    B. C. Lee and B. Y. Kim, "Development of an AI-Based Interview System for Remote Hiring," Int. J. Adv. Res. Eng. Technol., vol. 12, no. 3, pp. 654- 663, Mar. 2021, doi: 10.34218/IJARET.12.3.2021.060

    work page doi:10.34218/ijaret.12.3.2021.060 2021

  24. [24]

    Signaling theory: A review and assessment,

    B. L. Connelly, S. T. Certo, R. D. Ireland, and C. R. Reutzel, "Signaling theory: A review and assessment," J. Manage., vol. 37, no. 1, pp. 39 –67, Dec. 2010, doi: 10.1177/0149206310388419

    work page doi:10.1177/0149206310388419 2010

  25. [25]

    Liar, Liar, pants on fire: How verbal deception cues signal deceptive versus honest impression management and influence interview ratings,

    L. Roth, U. C. Klehe, and G. Willhardt, "Liar, Liar, pants on fire: How verbal deception cues signal deceptive versus honest impression management and influence interview ratings," Pers. Assessment Decis., vol. 7, no. 1, p. 7, May 2021, doi: 10.25035/pad.2021.01.007

    work page doi:10.25035/pad.2021.01.007 2021

  26. [26]

    Cues to deception,

    B. M. DePaulo, J. J. Lindsay, B. E. Malone, L. Muhlenbruck, K. Charlton, and H. Cooper, "Cues to deception," Psychological Bull., vol. 129, no. 1, pp. 74–118, Jan. 2003, doi: 10.1037/0033-2909.129.1.74

    work page doi:10.1037/0033-2909.129.1.74 2003

  27. [27]

    How fast are the leaked facial expressions: The duration of micro -expressions,

    W. J. Yan, Q. Wu, J. Liang, Y. H. Chen, and X. Fu, "How fast are the leaked facial expressions: The duration of micro -expressions," J. Nonverbal Behav., vol. 37 , no. 4, pp. 217 –230, Jul. 2013, doi: 10.1007/s10919-013-0159-8

    work page doi:10.1007/s10919-013-0159-8 2013

  28. [28]

    Automatic recognition of facial displays of unfelt emotions,

    K. Kulkarni et al., "Automatic recognition of facial displays of unfelt emotions," IEEE Trans. Affective Comput., vol. 12, no. 2, pp. 377–390, Apr. 2021, doi: 10.1109/taffc.2018.2874996

    work page doi:10.1109/taffc.2018.2874996 2021

  29. [29]

    A fake smile thwarts cheater detection,

    M. Okubo, A. Kobayashi, and K. Ishikawa, "A fake smile thwarts cheater detection," J. Nonverbal Behav., vol. 36, no. 3, pp. 217–225, Jul. 2012, doi: 10.1007/s10919-012-0134-9

    work page doi:10.1007/s10919-012-0134-9 2012

  30. [30]

    Darwin the detective: Observable facial muscle contractions reveal emotional high -stakes lies,

    L. ten Brinke, S. Porter, and A. Baker, "Darwin the detective: Observable facial muscle contractions reveal emotional high -stakes lies," Evolution Human Behav., vol. 33 , no. 4, pp. 411 –416, Jul. 2012, doi: 10.1016/j.evolhumbehav.2011.12.003

    work page doi:10.1016/j.evolhumbehav.2011.12.003 2012

  31. [31]

    Microexpressions differentiate truths from lies about future malicious intent,

    D. Matsumoto and H. C. Hwang, "Microexpressions differentiate truths from lies about future malicious intent," Frontiers Psychol., vol. 9, p. 2545, Dec. 2018, doi: 10.3389/fpsyg.2018.02545

    work page doi:10.3389/fpsyg.2018.02545 2018

  32. [32]

    Sparsity in dynamics of spontaneous subtle emotions: Analysis and application,

    A. C. Le Ngo, J. See, and R. C. W. Phan, "Sparsity in dynamics of spontaneous subtle emotions: Analysis and application," IEEE Trans. Affective Comput., vol. 8 , no. 3, pp. 396 –411, Jul. 2017, doi: 10.1109/taffc.2016.2523996

    work page doi:10.1109/taffc.2016.2523996 2017

  33. [33]

    Micro and macro facial expression recognition using advanced local motion patterns,

    B. Allaert, I. M. Bilasco, and C. Djeraba, "Micro and macro facial expression recognition using advanced local motion patterns," IEEE Trans. Affective Comput., vol. 13 , no. 1, pp. 147 –158, Oct. 2019, doi: 10.1109/taffc.2019.2949559

    work page doi:10.1109/taffc.2019.2949559 2019

  34. [34]

    Hiding true emotions: Micro -expressions in eyes retrospectively concealed by mouth movements,

    M. Iwasaki and Y. Noguchi, "Hiding true emotions: Micro -expressions in eyes retrospectively concealed by mouth movements," Sci. Rep., vol. 6, p. 22049, Feb. 2016, doi: 10.1038/srep22049

    work page doi:10.1038/srep22049 2016

  35. [35]

    A review of applicant faking in selection interviews,

    K. G. Melchers, N. Roulin, and A. K. Buehl, "A review of applicant faking in selection interviews," Int. J. Selection Assessment, vol. 28, no. 2, pp. 123–142, Feb. 2020, doi: 10.1111/ijsa.12280

    work page doi:10.1111/ijsa.12280 2020

  36. [36]

    Deception detection accuracy,

    J. K. Burgoon, "Deception detection accuracy," in The International Encyclopedia of Interpersonal Communication , C. R. Berger and M. E. Roloff, Eds. UK: WileyBlackwell, 2015, pp. 1–6

    work page 2015

  37. [37]

    Ekman, W

    P. Ekman, W. V. Friesen, and J. C. Hager, The Facial Action Coding System: A Technique for the Measurement of Facial Movement. Palo Alto, CA: Consulting Psychologists Press, 2002

    work page 2002

  38. [38]

    Microexpressions are not the best way to catch a liar,

    J. K. Burgoon, "Microexpressions are not the best way to catch a liar," Frontiers Psychol., vol. 9, p. 1672, Sept. 2018, doi: 10.3389/fpsyg.2018.01672

    work page doi:10.3389/fpsyg.2018.01672 2018

  39. [39]

    Exploring the movement dynamics of deception,

    N. D. Duran, R. Dale, C. T. Kello, C. N. H. Street, and D. C. Richardson, "Exploring the movement dynamics of deception," Frontiers Psychol., vol. 4, p. 140, Mar. 2013, doi: 10.3389/fpsyg.2013.00140

    work page doi:10.3389/fpsyg.2013.00140 2013

  40. [40]

    Deception detection in online automated job interviews,

    N. W. Twyman, S. J. Pentland, and L. Spitzley, "Deception detection in online automated job interviews," in HCI in Business, Government, and Organizations, F. F. H. Nah and B. S. Xiao, Eds. Cham, Switzerland: Springer International Publishing, 2018, pp. 206–216

    work page 2018

  41. [41]

    Intelligent deception detection through machine based interviewing,

    J. O’Shea, K. Crockett, W. Khan, P. Kindynis, A. Antoniades, and G. Boultadakis, "Intelligent deception detection through machine based interviewing," in 2018 Int. Joint Conf. Neural Netw. (IJCNN) , Rio de Janeiro, Brazil, Jul. 8-13 2018, pp. 1–8

    work page 2018

  42. [42]

    Silent talker: A new computer-based system for the analysis of facial cues to deception,

    J. Rothwell, Z. Bandar, J. O'Shea, and D. McLean, "Silent talker: A new computer-based system for the analysis of facial cues to deception," Appl. Cogn. Psychol., vol. 20 , no. 6, pp. 757 –777, Sept. 2006, doi: 10.1002/acp.1204

    work page doi:10.1002/acp.1204 2006

  43. [43]

    Lie detection based on facial micro expression body language and speech analysis,

    C. S. Barathi, “Lie detection based on facial micro expression body language and speech analysis,” Int. J. Eng. Res. Technol., vol. 5, pp. 337– 343, Feb. 2016, doi: 10.17577/IJERTV5IS020336

    work page doi:10.17577/ijertv5is020336 2016

  44. [44]

    Does “lie to me

    L. Su and M. Levine,“Does “lie to me” lie to you? An evaluation of facial clues to high-stakes deception,” Comp. Vis. Image Understand., vol. 147, pp. 52–68, Jun. 2016, doi: 10.1016/j.cviu.2016.01.009

    work page doi:10.1016/j.cviu.2016.01.009 2016

  45. [45]

    Deep facial expression recognition: A survey,

    S. Li and W. Deng, "Deep facial expression recognition: A survey," IEEE Trans. Affective Comput., vol. 13, no. 3, pp. 1195 –1215, Mar. 2020, doi: 10.1109/taffc.2020.2981446

    work page doi:10.1109/taffc.2020.2981446 2020

  46. [46]

    Measuring faking in the employment interview: Development and validation of an interview faking behavior scale,

    J. Levashina and M. A. Campion, "Measuring faking in the employment interview: Development and validation of an interview faking behavior scale," J. Appl. Psychol., vol. 92, no. 6, pp. 1638 –1656, Nov. 2007, doi: 10.1037/0021-9010.92.6.1638

    work page doi:10.1037/0021-9010.92.6.1638 2007

  47. [47]

    CAS(ME): A database for spontaneous macro-expression and micro-expression spotting and recognition,

    F. Qu, S. J. Wang, W. J. Yan, H. Li, S. Wu, and X. Fu, "CAS(ME): A database for spontaneous macro-expression and micro-expression spotting and recognition," IEEE Trans. Affective Comput., vol. 9, no. 4, pp. 424 – 436, Oct. 2018, doi: 10.1109/taffc.2017.2654440

    work page doi:10.1109/taffc.2017.2654440 2018

  48. [48]

    A survey of automatic facial micro-expression analysis: Databases, methods, and challenges,

    Y. H. Oh, J. See, A. C. Le Ngo, R. C. W. Phan, and V. M. Baskaran, "A survey of automatic facial micro-expression analysis: Databases, methods, and challenges," Frontiers Psychol., vol. 9, p. 1128, Jul. 2018, doi: 10.3389/fpsyg.2018.01128

    work page doi:10.3389/fpsyg.2018.01128 2018

  49. [49]

    [Online]

    Apple, Video recording, iPhone 14 - Technical Specifications, 2024. [Online]. Available: https://www.apple.com/iphone-14/specs/

    work page 2024

  50. [50]

    Motion profiles for deception detection using visual cues,

    N. Michael, M. Dilsizian, D. Metaxas, and J. K. Burgoon, "Motion profiles for deception detection using visual cues," in Computer vision – ECCV 2010, K. Daniilidis, P. Maragos, and N. Paragios, Eds. Berlin, Heidelberg: Springer, 2010, pp. 462–475

    work page 2010

  51. [51]

    [Online]

    Google, MediaPipe face mesh, MediaPipe, 2021. [Online]. Available: https://google.github.io/mediapipe/

    work page 2021

  52. [52]

    Real-time Facial Surface Geometry from Monocular Video on Mobile GPUs

    Y. Kartynnik, A. Ablavatski, I. Grishchenko, and M. Grundmann, "Real - time facial surface geometry from monocular video on mobile GPUs," arXiv:1907.06724, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1907

  53. [53]

    Survey, study and review of coordinate system notations using 3D environment for better understanding of EM fields behavior,

    K. S. Rana, M. L. Shaikh, and D. S. Ajij, "Survey, study and review of coordinate system notations using 3D environment for better understanding of EM fields behavior," in 2016 Thirteenth Int. Conf. Wireless Opt. Commun. Netw. (WOCN), Hyderabad, India, Jul. 21-23 2016, pp. 1–5

    work page 2016

  54. [54]

    3D convolutional neural network for machining feature recognition with gradient -based visual explanations from 3D CAD models,

    J. Lee, H. Lee, and D. Mun, "3D convolutional neural network for machining feature recognition with gradient -based visual explanations from 3D CAD models," Sci. Rep., vol. 12, no. 1, p. 14864, Sept. 2022, doi: 10.1038/s41598-022-19212-6

    work page doi:10.1038/s41598-022-19212-6 2022

  55. [55]

    Facial expression recognition using a temporal ensemble of multi -level convolutional neural networks,

    D. H. Nguyen, S. Kim, G. S. Lee, H. J. Yang, I. S. Na, and S. H. Kim, "Facial expression recognition using a temporal ensemble of multi -level convolutional neural networks," IEEE Trans. Affective Comput., vol. 13, pp. 226–237, Oct. 2019, doi: 10.1109/taffc.2019.2946540

    work page doi:10.1109/taffc.2019.2946540 2019

  56. [56]

    I. A. Essa and A. P. Pentland, 'Coding, analysis, interpretation, and recognition of facial expressions,' IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 7, pp. 757–763, Jul. 1997, doi: 10.1109/34.598232

    work page doi:10.1109/34.598232 1997

  57. [57]

    Human Emotion Recognition with Relational Region-Level Analysis,

    W. Li, X. Dong, and Y. Wang, "Human Emotion Recognition with Relational Region-Level Analysis," IEEE Trans. Affect. Comput., pp. 1, Mar. 2021, doi:10.1109/taffc.2021.3064918

    work page doi:10.1109/taffc.2021.3064918 2021

  58. [58]

    Deep learning model for real -time image compression in internet of underwater things (IoUT),

    N. Krishnaraj, M. Elhoseny, M. Thenmozhi, M. M. Selim, and K. Shankar, "Deep learning model for real -time image compression in internet of underwater things (IoUT)," J. Real-Time Image Process., vol. 17, no. 6, pp. 2097–2111, May 2019, doi: 10.1007/s11554-019-00879-6

    work page doi:10.1007/s11554-019-00879-6 2097

  59. [59]

    A 3D -CNN and LSTM based multi -task learning architecture for action recognition,

    X. Ouyang et al. , "A 3D -CNN and LSTM based multi -task learning architecture for action recognition," IEEE Access, vol. 7, pp. 40757–40770, Mar. 2019, doi: 10.1109/access.2019.2906654

    work page doi:10.1109/access.2019.2906654 2019

  60. [60]

    A CNN-based regression framework for estimating coal ash content on microscopic images,

    K. Zhang et al., "A CNN-based regression framework for estimating coal ash content on microscopic images," Measurement, vol. 189, p. 110589, Feb. 2022, doi: 10.1016/j.measurement.2021.110589

    work page doi:10.1016/j.measurement.2021.110589 2022

  61. [61]

    J. F. Hair, W. C. Black, B. J. Babin, and R. E. Anderson, Multivariate Data Analysis. EMEA, Hampshire, UK: Cengage Learning, 2019

    work page 2019

  62. [62]

    The use of cronbach’s alpha when developing and reporting research instruments in science education,

    K. S. Taber, "The use of cronbach’s alpha when developing and reporting research instruments in science education," Res. Sci. Educ., vol. 48, no. 6, pp. 1273–1296, Jun. 2018, doi: 10.1007/s11165-016-9602-2

    work page doi:10.1007/s11165-016-9602-2 2018

  63. [63]

    TensorFlow -based automatic personality recognition used in asynchronous video interviews,

    H. Y. Suen, K. E. Hung, and C. L. Lin, "TensorFlow -based automatic personality recognition used in asynchronous video interviews," IEEE Access, vol. 7, pp. 61018 –61023, Mar. 2019, doi: 10.1109/access.2019.2902863

    work page doi:10.1109/access.2019.2902863 2019

  64. [64]

    A model of faking likelihood in the employment interview,

    J. Levashina and M. A. Campion, "A model of faking likelihood in the employment interview," Int. J. Selection Assessment, vol. 14, no. 4, pp. 299–316, Nov. 2006, doi: 10.1111/j.1468-2389.2006.00353.x

    work page doi:10.1111/j.1468-2389.2006.00353.x 2006

  65. [65]

    Personality traits associated with various forms of lying,

    C. L. Hart, R. Lemon, D. A. Curtis, and J. D. Griffith, "Personality traits associated with various forms of lying," Psychological Stud., vol. 65, no. 3, pp. 239–246, Aug. 2020, doi: 10.1007/s12646-020-00563-x

    work page doi:10.1007/s12646-020-00563-x 2020

  66. [66]

    Spontaneous facial expression in unscripted social interactions can be measured automatically,

    J. M. Girard, J. F. Cohn, L. A. Jeni, M. A. Sayette, and F. De la Torre, "Spontaneous facial expression in unscripted social interactions can be measured automatically," Behav. Res. Methods, vol. 47, no. 4, pp. 1136 – 1147, Dec. 2015, doi: 10.3758/s13428-014-0536-1

    work page doi:10.3758/s13428-014-0536-1 2015

  67. [67]

    C. J. Kö nig, J. Wong, and G. Cen, 'How much do Chinese applicants fake?' Int. J. Sel. Assess., vol. 20, pp. 247–250, May. 2012, doi: 10.1111/j.1468- 2389.2012.00596

    work page doi:10.1111/j.1468- 2012

  68. [68]

    Shake and fake: The role of interview anxiety in deceptive impression management,

    D. M. Powell, J. S. Bourdage, and S. Bonaccio, "Shake and fake: The role of interview anxiety in deceptive impression management," J. Bus. Psychol., vol. 36, no. 5, pp. 829–840, Aug. 2021, doi: 10.1007/s10869-020- 09708-1

    work page doi:10.1007/s10869-020- 2021

  69. [69]

    Investigating the reliability of self -report data in the wild: The quest for ground truth,

    N. Gao, M. Saiedur Rahaman, W. Shao, and F. D. Salim, "Investigating the reliability of self -report data in the wild: The quest for ground truth," in Proc. Adjunct Proc. 2021 ACM Int. Joint Conf. Pervasive Ubiquitous Comput. Proc. 2021 ACM Int. Symp. Wearable Comput. , Sept. 2021, pp. 237–242. doi: 10.1145/3460418.3479338

    work page doi:10.1145/3460418.3479338 2021

  70. [70]

    A conceptual framework for investigating and mitigating machine learning measurement bias (MLMB) in psychological assessment,

    L. Tay, S. E. Woo, L. Hickman, B. M. Booth, and S. D'Mello, "A conceptual framework for investigating and mitigating machine learning measurement bias (MLMB) in psychological assessment," PsyArXiv, Jul. 2021, doi: 10.31234/osf.io/mjph3

    work page doi:10.31234/osf.io/mjph3 2021

  71. [71]

    A few transparent liars explaining 54% accuracy in deception detection experiments,

    T. R. Levine, "A few transparent liars explaining 54% accuracy in deception detection experiments," Ann. Int. Commun. Assoc., vol. 34, no. 1, pp. 41–61, May 2010, doi: 10.1080/23808985.2010.11679095

    work page doi:10.1080/23808985.2010.11679095 2010

  72. [72]

    Detecting deception through facial expressions in a dataset of videotaped interviews: A comparison between human judges and machine learning models,

    M. Monaro, S. Maldera, C. Scarpazza, G. Sartori, and N. Navarin, "Detecting deception through facial expressions in a dataset of videotaped interviews: A comparison between human judges and machine learning models," Comput. Human Behav., vol. 127, p. 107063, Feb. 2022, doi: 10.1016/j.chb.2021.107063. [74]X. Sun, Z. Pei, C. Zhang, G. Li, and J. Tao, 'Desig...

    work page doi:10.1016/j.chb.2021.107063 2022

  73. [73]

    J. L. Chen, M. Li, M. Wu, W. Pedrycz, and K. Hirota, 'Convolutional Features-Based Broad Learning With LSTM for Multidimensional Facial Emotion Recognition in Human –Robot Interaction,' IEEE Trans. Syst., Man, Cybern. Syst ., vol. 54, no. 1, pp. 64 -75, Jan. 2024, doi: 10.1109/TSMC.2023.3301001. Hung-Yue Suen is a Distinguished Associate Professor of Tech...

    work page doi:10.1109/tsmc.2023.3301001 2024