Pith. sign in

REVIEW 4 major objections 4 minor 48 references

Experimental Evidence That AI-Managed Workers Tolerate Lower Pay Without Demotivation

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper reports that a rule-based AI manager rated workers more harshly and paid 40% less than human managers, yet those workers felt the treatment was fair and stayed motivated.

desk verdict A genuinely novel behavioral experiment showing AI evaluation can flatten fairness reactions to low pay, with a caveat about a lenient human-manager baseline. read the letter →

arxiv 2505.21752 v1 pith:5VFHWFS4 submitted 2025-05-27 cs.CY cs.HC

Mengchen Dong , Levin Brinkmann , Omar Sherif , Shihan Wang , Xinyu Zhang , Jean-François Bonnefon , Iyad Rahwan This is my paper
classification cs.CYcs.HC
keywords algorithmicmanagementAIevaluationworkermotivationfairnessperceptionswagesettingMinecraftexperimentbehavioralsilentexploitation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper asks whether workers will accept lower pay from an AI manager without losing motivation, and answers with a controlled but richly realistic experiment in a custom Minecraft workplace. Three hundred eighty-two workers completed repeated iron-pickaxe production tasks under human, rule-based AI, decision-tree AI, or AI-advised human management, with wages tied to each evaluation. The paper reports that the rule-based AI manager, trained on human-defined evaluation principles, gave lower ratings and paid 40% less per round than human managers, yet these workers still felt the treatment was fair and stayed motivated. The key mechanism is that fairness perceptions under human management tracked whether the evaluation matched workers' self-assessed skill, while under this AI manager that link was much weaker, so low evaluations no longer triggered an emotional response. A second, harsher AI that cut wages by 60% did damage fairness and motivation, showing that the quiet acceptance of lower pay has boundaries.

What carries the argument

The load-bearing object is the discrepancy between a worker's self-assessed competence and the manager's evaluation, and the sensitivity of fairness perceptions to that discrepancy. Workers rated their Minecraft skill as beginner, intermediate, or advanced in advance, and were then evaluated on the same three-level scale; multilevel regressions compare how fairness judgments respond to being rated worse, the same, or better than one's self-assessment. The critical result is an interaction: under human management this fairness–discrepancy link is strong, while under the rule-based AI manager it is significantly flattened. Because next-round motivation is carried by perceived fairness regardless of manager type, flattening the fairness link is what allows wages to drop by 40% without a motivational cost. The proposed psychological mechanism is the perceived impartiality and lack of intentionality of algorithmic evaluation, which weakens the emotional reaction that normally accompanies a downgrade from a human.

What would settle it

Re-run the experiment with human managers randomly selected rather than chosen from top performers, and pay them for rating accuracy instead of a flat bonus; if their ratings become as harsh as the rule-based AI's (modal 'intermediate', roughly 40% lower wages), the wage gap is explained by manager selection and incentives rather than by the AI-ness of the evaluator.

Watch

Extended reading notes

Core claim

Workers (N = 382) self-assessed as beginner, intermediate, or advanced before each session, then received an evaluation on the same scale and a wage of ¢0, ¢20, or ¢50 after each 10-minute attempt. Human managers most often awarded 'advanced' (46%), whereas the rule-based AI (AI-R) most often awarded 'intermediate' (88%) and the decision-tree AI (AI-T) split between 'beginner' and 'intermediate'. Average wages fell from 29 cents per round under human management to 18 cents under AI-R (a 40% reduction) and 11 cents under AI-T (a 60% reduction). Under human management, perceived fairness rose when evaluations matched or exceeded self-assessments and fell when evaluations were worse, but under AI-R this sensitivity to discrepancy was significantly flattened. Since motivation in the following round tracked perceived fairness regardless of manager type, AI-R workers stayed motivated despite lower pay; AI-T workers, whose fairness perceptions dropped overall, did not. In the AI-advised human condition, fairness sensitivity reappeared, supporting the paper's account that the emotional response to being downgraded is muted specifically when the evaluator is an algorithm perceived as impartial and lacking intent. The paper calls the outcome 'silent exploitation.'

Load-bearing premise

The comparison assumes that the human managers, who were top performers from the training phase and were paid a flat bonus per round regardless of rating accuracy, provide an unbiased baseline for human management; if their leniency comes from selection or incentives rather than from being human, the 40% wage gap is not purely an AI effect.

Editorial extensions

If this is right

  • If the central result holds, organizations can deploy rule-based AI managers to reduce wage costs without a detectable drop in worker morale, because low evaluations no longer register as unfair.
  • Worker self-reports of fairness and motivation are not a reliable early-warning system for AI-driven wage cuts; the paper's findings imply that the distributions of ratings and wages themselves need to be audited.
  • Delivering AI-generated evaluations through a human restores the fairness–discrepancy link, so keeping a human in the loop may re-arm the social reactions that constrain extractive pay practices.
  • The quiet-acceptance effect has a boundary: a harsher AI that cut wages by 60% reduced fairness perceptions and motivation, so not every AI system can cut pay without backlash.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not test whether telling workers that the AI has discretion or can err would restore fairness sensitivity; if perceived intentionality is the active ingredient, that manipulation should bring the emotional response back.
  • One implication the authors leave implicit is that conventional fairness surveys may systematically under-report exploitation under AI management, because the muted response means workers feel less aggrieved even when pay is objectively lower.
  • The same muted-reaction mechanism might generalize beyond employment to algorithmic pricing, automated fines, or AI-administered benefits, where AI errors could generate less grievance than equivalent human errors; that is an extrapolation, not a result of this paper.
  • A longer time horizon could erode the effect as workers learn the AI's rating distribution; the three-round design cannot tell whether silent exploitation persists in long-term employment relationships.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper presents a randomized experiment (N = 382 workers) in a custom Minecraft workplace, the Iron Pickaxe Factory, to compare human, AI, and hybrid management. Workers completed three rounds of an iron pickaxe task; each round was followed by a manager evaluation (beginner/intermediate/advanced) and contingent pay, and workers reported fairness and motivation. A rule-based AI manager (AI-R) trained on human-defined principles assigned lower ratings and paid 40% less than human managers, yet did not reduce perceived fairness or subsequent motivation. In contrast, a decision-tree AI (AI-T) paid 60% less and did reduce fairness and motivation. The authors argue that fairness perceptions under human management tracked the gap between self-assessed and manager-assigned ratings, while this sensitivity was muted under AI-R, and that motivation tracked fairness, so AI-managed workers remained motivated despite lower pay.

Significance. The study is methodologically ambitious and a valuable advance over vignette studies: it uses a functioning Minecraft workplace with real-time monitoring, repeated evaluation-feedback rounds, contingent wages, and genuine AI evaluation systems rather than imagined AI. The use of multilevel models with confidence intervals, simulation-based power analysis, and a pre-experimental self-assessment baseline are strengths. If the causal contrast is valid, the finding that AI evaluation suppresses the fairness backlash to downgrades—and that a human delivering AI advice restores it—is an important, policy-relevant result for algorithmic management research. The main risks are the internal validity of the human-manager baseline and the incomplete documentation of the AI systems; these do not undermine the experimental design but do determine the interpretation of the headline wage effect.

major comments (4)
  1. [Methods (Design and procedure); Results (Manager evaluations)] The human-manager baseline is not exchangeable with the AI conditions. Human managers were the participants who passed the training session first and were paid a flat $0.5 per round with no accuracy incentive, and their modal rating was 'advanced' (46%), while workers' self-assessments were only 31% advanced. Since the Discussion states there is 'no clear ground truth' about performance, the 40% wage reduction under AI-R is measured against a selected, unincentivized, and lenient baseline. The Human+AI condition does not fix this, because those managers were free to ignore the AI recommendation and their wages stayed closer to the Human condition than to AI-R. Please address selection and incentives directly (e.g., compare manager ratings by training performance, add an accuracy bonus, or report a robustness analysis with a different baseline).
  2. [Methods (Design and procedure) and Discussion] The AI-R system is under-specified. Methods says it was 'trained on human-defined evaluation principles' based on pace of tech-tree advancement, while Discussion adds it was 'fitted to the distribution of human evaluation data.' The principles, fitting data, thresholds, and feature-to-rating mapping are not reported. This is load-bearing because AI-R is the main treatment and its harshness is the proposed AI effect; without a specification, the reader cannot tell whether the model's strictness is a property of the implemented algorithm or of arbitrary thresholds. The AI-T decision tree is similarly described only by ethics approval numbers. Please provide the full algorithm specification or make code and parameters available.
  3. [Methods (Participants) and Results (Motivation)] The claim that AI-R workers maintained motivation rests on a null effect (β = -0.07, 95% CI [-0.20, 0.05]). The reported power analysis targets the fairness outcome, not the motivation outcome. To treat 'without demotivation' as supported evidence, the authors should report an equivalence or sensitivity analysis for the motivation contrast, or pre-specify a minimal meaningful decline and show the CI excludes it.
  4. [Abstract and Results (Fairness perceptions)] The abstract says the effects were 'driven by a muted emotional response to AI evaluation,' but no emotion measure appears in the Methods. The observed condition-by-discrepancy interaction on fairness is consistent with the emotional account, but it does not distinguish it from alternatives such as lower expectations for AI or different attribution of intent. Please either add an emotion measure or soften the causal mechanism claim in the abstract and Discussion.
minor comments (4)
  1. [Figure 2 caption] The caption uses '$0.2' and '$0.5' while the text uses '¢20' and '¢50'; please standardize the currency notation.
  2. [Results (Manager evaluations)] The Human modal rating percentages (23%, 32%, 46%) sum to 101%; please correct the rounding.
  3. [Methods (Participants)] The total manager N is reported as 102, but the two condition sizes (52 + 52) sum to 104; please correct this numerical inconsistency.
  4. [Methods (Design and procedure)] The motivation item labeled 'nervosity' should be 'nervousness' to match standard English usage and the likely source scale.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation: the wage gap and fairness effects are measured experimental outcomes, not fitted predictions.

full rationale

The paper contains no mathematical derivation that reduces to its own inputs. The AI-R and AI-T managers are treatment manipulations: their evaluation rules are fixed before outcomes are observed, and the 40% wage reduction, the flattened fairness-discrepancy slope, and the preserved motivation are estimated from participant responses rather than produced by fitting parameters to those outcomes. The only fitting mentioned—the rule-based model 'was fitted to the distribution of human evaluation data'—refers to calibrating the AI treatment, not to predicting the study's dependent variables; the empirical contrast with the concurrently measured Human manager condition is not definitional. Self-citations to Dong, Bonnefon, and Rahwan (2024) appear in the Introduction as background and in Methods as the source of an adapted motivation scale; neither is load-bearing. The Discussion's explicit limitations (one AI system tested, disabled communication, short duration) are honest scope caveats, not admissions of circularity. A skeptical concern about the human-manager baseline's leniency due to selection and flat bonuses is a validity or confound issue, not circularity, and does not raise the circularity score.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

No new physical or conceptual entities are introduced. The closest things to free parameters are the unreported parameters of the two AI managers, which are treatment inputs rather than fitted outcomes. The axioms are domain assumptions about task validity, measurement validity, and comparability of conditions.

free parameters (2)
  • AI-R evaluation thresholds = unreported
    Rule-based AI uses thresholds on pace of advancement through the tech tree; these determine the rating distribution (11% beginner, 88% intermediate, 1% advanced) and hence the 40% wage cut. Values are not given and the model was fitted to human evaluation data.
  • AI-T decision-tree parameters = unreported
    Trained on crowdworker ratings of gameplay videos; parameters and training details are not reported. This system produced a 60% wage reduction and backlash.
assumptions (5)
  • domain assumption Workers' self-assessed competence is a meaningful benchmark for expected evaluations and wages.
    The discrepancy between self-assessment and manager rating is the main independent variable in the fairness and wage-deviation analyses (Fig. 2A, 2B, 2D). If self-assessments are systematically biased, the asymmetry in response is harder to interpret.
  • domain assumption Self-reported Likert items capture motivation and fairness as intended.
    Motivation (enjoyment, effort, reverse-coded pressure, autonomy) and fairness (procedural and distributive) are measured with brief scales; no behavioral productivity measure is used to validate the motivation construct.
  • domain assumption The Minecraft iron-pickaxe task is an ecologically valid analog of real production work.
    The paper's external-validity argument rests on task autonomy, real-time monitoring, and repeated rounds; the authors acknowledge generalization limits in the Discussion.
  • domain assumption Random assignment and the shared interface make management condition the only systematic difference between groups.
    Communication was disabled in all conditions and both human and AI conditions showed identical avatars; this is a deliberate design choice, but it also removes the social context of real workplaces.
  • standard math Multilevel regression assumptions hold for the nested worker-round data.
    Statistical claims rely on standard mixed-model inference; no robust checks are reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Experimental Evidence That AI-Managed Workers Tolerate Lower Pay Without Demotivation." pith.science (2026). https://pith.science/paper/5VFHWFS4

@misc{pith2026250521752,
  author       = {Pith},
  title        = {Pith review of: Experimental Evidence That AI-Managed Workers Tolerate Lower Pay Without Demotivation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5VFHWFS4}},
  note         = {Machine review of arXiv:2505.21752}
}
read the original abstract

Experimental evidence on worker responses to AI management remains mixed, partly due to limitations in experimental fidelity. We address these limitations with a customized workplace in the Minecraft platform, enabling high-resolution behavioral tracking of autonomous task execution, and ensuring that participants approach the task with well-formed expectations about their own competence. Workers (N = 382) completed repeated production tasks under either human, AI, or hybrid management. An AI manager trained on human-defined evaluation principles systematically assigned lower performance ratings and reduced wages by 40\%, without adverse effects on worker motivation and sense of fairness. These effects were driven by a muted emotional response to AI evaluation, compared to evaluation by a human. The very features that make AI appear impartial may also facilitate silent exploitation, by suppressing the social reactions that normally constrain extractive practices in human-managed work.

Figures

Figures reproduced from arXiv: 2505.21752 by the authors.

Figure 1
Figure 1. The experiment interface. An example of the in￾terface for workers under AI management. Workers com￾plete the iron pickaxe task for three repeated rounds, each lasting 10 minutes. With this experimental setup, we directly compare the ef￾fects of human versus AI management on worker evalua￾tions, wages, fairness perceptions, and motivational dynam￾ics. Here we show that an AI management system trained on human-define… view at source ↗
Figure 2
Figure 2. Summary of the main results (worker N = 382). Both self-assessment and manager evaluation were implemented at three levels: beginner ($0), intermediate ($0.2), and advanced ($0.5). Each level corresponds to a different bonus, as shown in the parentheses. Fairness scores aggregate procedural and distributive fairness; segmenting the two dimensions yielded similar results. management, fairness perceptions closely trac… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

48 extracted references · 41 canonical work pages

  1. [1]

    H.; Compagnone, M.; and Laske, M

    Acikgoz, Y.; Davison, K. H.; Compagnone, M.; and Laske, M. 2020. Justice perceptions of artificial intelligence in selection. International Journal of Selection and Assessment, 28: 399--416

    work page 2020

  2. [2]

    Amresh, A.; Cooke, N.; and Fouse, A. 2023. A Minecraft Based Simulated Task Environment for Human AI Teaming. In Proceedings of the 23rd ACM International Conference on Intelligent Virtual Agents, 1--3. Würzburg, Germany: ACM

    work page 2023

  3. [3]

    Anteby, M.; and Chan, C. K. 2018. A self-fulfilling cycle of coercive surveillance: Workers’ invisibility practices and managerial justification. Organization Science, 29: 247--263

    work page 2018

  4. [4]

    Bai, B.; Dai, H.; Zhang, D.; Zhang, F.; and Hu, H. 2020. The Impacts of Algorithmic Work Assignment on Fairness Perceptions and Productivity: Evidence from Field Experiments. SSRN Electronic Journal, 1--32

    work page 2020

  5. [5]

    J.; Bashshur, M

    Barclay, L. J.; Bashshur, M. R.; and Fortin, M. 2017. Motivated cognition and fairness: Insights, integration, and creating a path forward. Journal of Applied Psychology, 102: 867--889

    work page 2017

  6. [6]

    M.; and Jentsch, F

    Bendell, R.; Williams, J.; Fiore, S. M.; and Jentsch, F. 2024. Individual and team profiling to support theory of mind in artificial social intelligence. Scientific Reports, 14: 12635

    work page 2024

  7. [7]

    E.; Wilson, D.; Arnestad, M

    Bigman, Y. E.; Wilson, D.; Arnestad, M. N.; Waytz, A.; and Gray, K. 2023. Algorithmic discrimination causes less moral outrage than human discrimination. Journal of Experimental Psychology: General, 152: 4--27

    work page 2023

  8. [8]

    L.; Schou, P

    Bucher, E. L.; Schou, P. K.; and Waldkirch, M. 2021. Pacifying the algorithm – Anticipatory compliance in the face of algorithmic management in the gig economy. Organization, 28: 44--67

    work page 2021

Show all 48 references
  1. [9]

    D.; and Rahman, H

    Cameron, L. D.; and Rahman, H. 2022. Expanding the Locus of Resistance: Understanding the Co-constitution of Control and Resistance in the Gig Economy. Organization Science, 33: 38--58

    2022

  2. [10]

    S.; De Cremer, D.; Oh, E.-J.; and Kim, Y

    Chun, J. S.; De Cremer, D.; Oh, E.-J.; and Kim, Y. 2024. What algorithmic evaluation fails to deliver: respectful treatment and individualized consideration. Scientific Reports, 14: 25996

    2024

  3. [11]

    Cohen-Charash, Y.; and Spector, P. E. 2001. The Role of Justice in Organizations: A Meta-Analysis. Organizational Behavior and Human Decision Processes, 86: 278--321

    2001

  4. [12]

    Colquitt, J. A. 2001. On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86: 386--400

    2001

  5. [13]

    Crawford, K. 2021. Atlas of AI: The Real Worlds of Artificial Intelligence. Yale University Press

    2021

  6. [14]

    R.; and Dunning, D

    Critcher, C. R.; and Dunning, D. 2009. How chronic self-views influence (and mislead) self-assessments of task performance: Self-views shape bottom-up experiences with the task. Journal of Personality and Social Psychology, 97: 931--945

    2009

  7. [15]

    L.; Eghrari, H.; Patrick, B

    Deci, E. L.; Eghrari, H.; Patrick, B. C.; and Leone, D. R. 1994. Facilitating Internalization: The Self‐Determination Theory Perspective. Journal of Personality, 62: 119--142

    1994

  8. [16]

    Delfanti, A. 2021. Machinic dispossession and augmented despotism: Digital work in an Amazon warehouse. New Media & Society, 23: 39--55

    2021

  9. [17]

    S.; and Ter Haar, B

    Doellgast, V.; Litwin, A. S.; and Ter Haar, B. 2021. AI in Contact Centers: Artificial Intelligence and Algorithmic Management in Frontline Service Workplaces. https://hdl.handle.net/1813/113706

    2021

  10. [18]

    Doellgast, V.; Wagner, I.; and O’Brady, S. 2023. Negotiating limits on algorithmic management in digitalised services: cases from Germany and Norway. Transfer: European Review of Labour and Research, 29: 105--120

    2023

  11. [19]

    Dong, M.; Bonnefon, J.-F.; and Rahwan, I. 2024. Toward human-centered AI management: Methodological challenges and future directions. Technovation, 131: 102953

    2024

  12. [20]

    Duani, N.; Barasch, A.; and Morwitz, V. 2024. Demographic Pricing in the Digital Age: Assessing Fairness Perceptions in Algorithmic versus Human-Based Price Discrimination. Journal of the Association for Consumer Research, 9: 257--268

    2024

  13. [21]

    M.; Kim, T.; and Duhachek, A

    Garvey, A. M.; Kim, T.; and Duhachek, A. 2023. Bad News? Send an AI. Good News? Send a Human. Journal of Marketing, 87: 10--25

    2023

  14. [22]

    H.; and et al

    Guss, W. H.; and et al. 2021. The MineRL 2020 Competition on Sample Efficient Reinforcement Learning using Human Priors. https://doi.org/10.48550/ARXIV.2101.11071

    2021 doi arXiv

  15. [23]

    Hafner, D.; Pasukonis, J.; Ba, J.; and Lillicrap, T. 2025. Mastering diverse control tasks through world models. Nature

    2025

  16. [24]

    S.; and Laurin, K

    Jago, A. S.; and Laurin, K. 2022. Assumptions About Algorithms’ Capacity for Discrimination. Personality and Social Psychology Bulletin, 48: 582--595

    2022

  17. [25]

    T.; and Naaman, M

    Jakesch, M.; Hancock, J. T.; and Naaman, M. 2023. Human heuristics for AI-generated language are flawed. Proceedings of the National Academy of Sciences, 120: e2208839120

    2023

  18. [26]

    C.; Valentine, M

    Kellogg, K. C.; Valentine, M. A.; and Christin, A. 2020. Algorithms at work: The new contested terrain of control. Academy of Management Annals, 14: 366--410

    2020

  19. [27]

    Lafit, G.; and et al. 2021. Selection of the Number of Participants in Intensive Longitudinal Studies: A User-Friendly Shiny App and Tutorial for Performing Power Analysis in Multilevel Regression Models That Account for Temporal Dependencies. Advances in Methods and Practices...

    2021

  20. [28]

    J.; and Elliot, R

    Lawler, J. J.; and Elliot, R. 1996. Artificial Intelligence in HRM: An Experimental Study of an Expert System. Journal of Management, 22: 85--111

    1996

  21. [29]

    Lee, M. K. 2018. Understanding perception of algorithmic decisions: Fairness, trust, and emotion in response to algorithmic management. Big Data & Society, 5: 1--16

    2018

  22. [30]

    Liu, Y.; and et al. 2024. Aligning Cyber Space with Physical World: A Comprehensive Survey on Embodied AI. https://doi.org/10.48550/ARXIV.2407.06886

    2024 doi arXiv

  23. [31]

    Mohlmann, M.; and Zalmanson, L. 2018. Hands on the Wheel: Navigating Algorithmic Management and Uber Drivers’ Autonomy. In ICIS 2017: Transforming Society through Digital Innovation

    2018

  24. [32]

    Muldoon, J.; Cant, C.; Graham, M.; and Ustek Spilda, F. 2023. The poverty of ethical AI: impact sourcing and AI supply chains. AI & Society

    2023

  25. [33]

    T.; Fast, N

    Newman, D. T.; Fast, N. J.; and Harmon, D. J. 2020. When eliminating bias isn’t fair: Algorithmic reductionism and procedural justice in human resource decisions. Organizational Behavior and Human Decision Processes, 160: 149--167

    2020

  26. [34]

    Peters, H.; Kyngdon, A.; and Stillwell, D. 2021. Construction and validation of a game-based intelligence assessment in minecraft. Computers in Human Behavior, 119: 106701

    2021

  27. [35]

    Rahman, H. A. 2021. The Invisible Cage: Workers’ Reactivity to Opaque Algorithmic Evaluations. Administrative Science Quarterly, 66: 945--988

    2021

  28. [36]

    Ranganathan, A.; and Benson, A. 2020. A Numbers Game: Quantification of Work, Auto-Gamification, and Worker Productivity. American Sociological Review, 85: 573--609

    2020

  29. [37]

    Schlund, R.; and Zitek, E. M. 2024. Algorithmic versus human surveillance leads to lower perceptions of autonomy and increased resistance. Communications Psychology, 2: 53

    2024

  30. [38]

    C.; and et al

    Simon, K. C.; and et al. 2022. Sleep facilitates spatial memory but not navigation using the Minecraft Memory and Navigation task. Proceedings of the National Academy of Sciences, 119: e2202394119

    2022

  31. [39]

    G.; and Bauer, K

    Sitzmann, T.; Ely, K.; Brown, K. G.; and Bauer, K. N. 2010. Self-Assessment of Knowledge: A Cognitive Learning or Affective Measure? Academy of Management Learning & Education, 9: 169--191

    2010

  32. [40]

    J.; Hu, H.; and Van Mieghem, J

    Sun, J.; Zhang, D. J.; Hu, H.; and Van Mieghem, J. A. 2021. Predicting Human Discretion to Adjust Algorithmic Prescription: A Large-Scale Field Experiment in Warehouse Operations. Management Science

    2021

  33. [41]

    Tong, S.; Jia, N.; Luo, X.; and Fang, Z. 2021. The Janus face of artificial intelligence feedback: Deployment versus disclosure effects on employee performance. Strategic Management Journal, 42: 1600--1631

    2021

  34. [42]

    Wang, G.; and et al. 2023. Voyager: An Open-Ended Embodied Agent with Large Language Models. https://doi.org/10.48550/ARXIV.2305.16291

    2023 doi arXiv

  35. [43]

    Wood, A. 2021. Algorithmic Management: Consequences for Work Organisation and Working Conditions. https://publications.jrc.ec.europa.eu/repository/handle/JRC124874

    2021

  36. [44]

    Xing, W.; Li, M.; Li, M.; and Han, M. 2025. Towards Robust and Secure Embodied AI: A Survey on Vulnerabilities and Attacks. https://doi.org/10.48550/ARXIV.2502.13175

    2025 doi arXiv

  37. [45]

    Yalcin, G.; Lim, S.; Puntoni, S.; and Van Osselaer, S. M. J. 2022. Thumbs Up or Down: Consumer Reactions to Decisions by Algorithms Versus Humans. Journal of Marketing Research, 59: 696--717

    2022

  38. [46]

    Yaraghi, N.; Ramesh, R.; and Tayi, G. K. 2024. Pay, Pat, and Clawback: Incentivizing Service Providers’ Participation in On-Demand Digital Platforms. Management Science

    2024

  39. [47]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  40. [48]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.