Pith. sign in

REVIEW 5 major objections 6 minor 74 references

Let people fail! Exploring the influence of explainable virtual and robotic agents in learning-by-doing tasks

T0 review · 5 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper claims that in a learning-by-doing task, participants who learned without any explainable AI agent explored more and ended up with better task knowledge than those assisted by a computer or a humanoid robot.

desk verdict Solid empirical study with a provocative counterintuitive result, but the central interpretation rests on an unvalidated proxy for exploration. read the letter →

arxiv 2411.10176 v1 pith:7GPBSEYB submitted 2024-11-15 cs.AI cs.HCcs.RO

Marco Matarese , Francesco Rea , Katharina J. Rohlfing , Alessandra Sciutti This is my paper · ORCID
classification cs.AIcs.HCcs.RO
keywords explainableartificialintelligencehuman-robotinteractionlearning-by-doinghuman-AIcollaborationautomationbiascontrastiveexplanationsuserstudyrobotictutoring
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

This paper asks whether explainable AI agents actually help people learn a new task, and reports a counterintuitive result: the people who received no help at all learned more. In a simulated nuclear-power-plant management task, participants who worked alone made more mistakes during training, but those mistakes are read as exploration, and the self-taught group went on to score significantly higher on a post-task knowledge test than participants assisted by an explainable computer or humanoid robot. The paper also found that two explanation styles, classical and partner-aware contrastive explanations, produced no difference in final knowledge, but did change behavior: with a computer, adaptive explanations made people faster, while with a robot they made people more likely to follow its suggestions. The authors conclude that explainable assistance can reduce learners' agency and exploration, and suggest that automated tutors should invite people to fail and explore rather than simply follow suggestions.

What carries the argument

The central object is the learning-by-doing assessment task itself: a simulated nuclear power plant with hidden rules that participants must discover by acting, with a 30-minute training phase, a 10-minute assessment phase, and a final knowledge test. The expert agents are driven by a deterministic decision tree trained with the Conservative Q-Improvement reinforcement-learning algorithm, which can answer what and why questions by tracing splits. The two explanation strategies are classical selection of the most relevant feature and partner-aware contrastive selection that compares the agent's suggestion with the user's predicted action. The argument's load-bearing measure is the number of anomalies (conditions that damage the simulated plant) during training, which the paper treats as an estimator of exploration.

What would settle it

Conduct the same learning-by-doing task with a fixed number of training steps instead of a fixed time limit; if the unassisted group no longer shows more anomalies and no longer outscores the assisted groups, the exploration explanation fails. Alternatively, measure exploration directly, for example by counting distinct environment states visited or computing information gain, and check whether that measure, rather than the anomaly count, drives the test-score difference.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that assisted learning underperforms unassisted learning in a learning-by-doing task. In a between-subjects study, the Self-taught group produced more actions, more energy, more critical steps, and more anomalies during training than all four assisted groups, and then outperformed the COM C-XAI, COM A-XAI, and Robot A-XAI groups on the post-experiment test (ANOVA F(4)=3.99, p=.007), with the effect concentrated in scenario questions (F(4)=6.37, p<.001). The authors interpret the higher anomaly count as evidence of deeper exploration, and argue that interaction with expert explainable agents encouraged over-reliance and reduced autonomous exploration, leaving assisted participants with weaker ability to generalize.

Load-bearing premise

The paper's argument depends on treating the number of anomalies during training as a good estimator of exploration, and on assuming that exploration, not some other factor like time pressure or prior knowledge, is what gave the unassisted group its higher test scores.

Editorial extensions

If this is right

  • If correct, explainable agents that merely answer questions can reduce learning gain in learning-by-doing tasks, even when they improve immediate performance.
  • Automated tutoring systems should be designed to encourage autonomous exploration, tolerating errors during training as a price for deeper understanding.
  • Partner-aware contrastive explanations did not improve final knowledge over classical explanations, but they did change behavior: faster decisions with a computer and more compliance with a robot.
  • The self-taught group's advantage appeared specifically in scenario-based generalization questions, suggesting that unassisted exploration builds transferable causal knowledge rather than rote procedures.

Reading between the lines

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

  • Beyond the paper: one could test whether adding a mechanism that forces users to commit to a prediction before seeing the agent's suggestion closes the gap between assisted and unassisted learners, since the paper identifies over-reliance as the likely culprit.
  • Beyond the paper: the different effects of the same adaptive explanation strategy on computer versus robot interaction suggest that embodiment and social presence may change whether explanations promote reflection or compliance; this could be isolated by matching the agents' wording and nonverbal cues more tightly.
  • Beyond the paper: the anomaly-as-exploration proxy deserves direct validation; if validated, it gives XAI researchers a simple behavioral marker for exploration in learning tasks, independent of self-reports.
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

5 major / 6 minor

Summary. This paper reports a between-subject user study (N = 55) on learning-by-doing with a simulated nuclear power plant task. Participants either worked autonomously (Self-taught, n = 11) or were assisted by a computer or the iCub robot that could answer 'what' and 'why' queries using either classical (C-XAI) or partner-aware contrastive (A-XAI) explanations derived from a deterministic decision-tree policy. The authors compare behavior in a 30-minute training phase and a 10-minute assessment phase, together with a post-experiment knowledge test. They find that A-XAI made computer users move faster, made the robot more persuasive after 'why' questions, and had no significant effect on final task knowledge. The headline result is that Self-taught participants performed more actions, produced more energy and more anomalies, and scored higher on the post-experiment test, especially on scenario questions, than most assisted groups. The authors interpret the anomaly count as a measure of exploration and argue that unassisted 'failing' and exploring produced better learning, thereby rejecting their hypothesis H2.

Significance. The study's strengths are its direct measurement of learning via a post-experiment knowledge test, its inclusion of a no-assistance baseline, and its concrete manipulation of explanation strategies through a deterministic interpretable decision tree. If the headline effect is robust, it has practical implications for automated tutoring and for designing AI assistants that scaffold rather than replace exploration. The paper also gives useful, explicit credit to the possibility that over-reliance or automation bias explains the assisted groups' lower test scores. However, the central interpretation depends on an unvalidated proxy (anomalies as exploration) and is entangled with a practice-volume confound; in addition, several statistical and reporting issues need correction. With those issues addressed, this could be a valuable contribution to HCI and XAI research.

major comments (5)
  1. [Section 4.6.2 and Section 5] The conclusion that the Self-taught group's higher post-test scores were caused by greater 'exploration' rests entirely on the assertion in Section 5 that 'the number of anomalies during training is a good estimator of participants' degree of exploration.' This assertion is not secured: anomalies are damaging outcomes, not a validated measure of exploration, and no independent index such as state-space coverage, action diversity, information gain, or query behavior is reported. Because the Self-taught group also performed significantly more actions (Section 4.6.1), the anomaly count is confounded with practice volume. Without a per-action anomaly rate or a mediation analysis linking anomalies to test scores, the data are equally consistent with the alternative explanation that more hands-on practice, rather than autonomous exploration or failure-driven learning, produced the test advantage.
  2. [Sections 3.1, 4.6.1, and 5] The fixed-time design directly creates the practice-volume confound: Self-taught participants did not spend time querying an agent, so they could perform more actions in the same 30 minutes. The manuscript acknowledges in the limitations paragraph that it is 'unclear whether we would observe the same results by removing or changing such a constraint, i.e., using a fixed number of steps,' but it does not test this. Since the behavioral differences (actions, energy, anomalies, critic steps) and possibly the test differences depend on this design choice, this limitation is load-bearing rather than peripheral.
  3. [Section 4.1 and Section 5] The baseline knowledge check is only marginally non-significant (chi-square(8) = 14.1, p = .078), and the authors themselves note a 'slight disparity' between the COM group (Germany) and the Self-taught and Robot groups (Italy). The cross-group ANOVAs in Sections 4.6.1 and 4.6.2 do not include site or prior knowledge as a covariate, and the statement that there were 'no differences between these two macro-groups regarding performance in training and assessment' is made without a supporting test. This leaves open the possibility that the Self-taught group's superior test performance partly reflects pre-existing differences rather than the absence of assistance.
  4. [Section 4.2 and Table 1] The manipulation check in Table 1 is based on sixteen independent-samples t-tests without any multiple-comparison correction, so the reported 'five out of eight' and 'four out of eight' significant explanandum differences are not strong evidence that the two explanation strategies systematically differed. Uncorrected t-tests also appear in Sections 4.3-4.5 (for example, t = 2.21, p = .039, and t = 2.226, p = .038). Please report adjusted p-values or explicitly justify these comparisons as exploratory.
  5. [Section 4.5 and Figure 8] Reporting the same t-statistic (t = 2.226) and p-value (p = .038) for both the COM and Robot subgroups in the comparison of 'Follow AI' percentages during why-questions is internally inconsistent if this is a single between-group independent-samples test. As written, the reader cannot tell whether this is a paired comparison, a typographical error, or a duplicated statistic. Please clarify the test, report degrees of freedom and effect size, and correct the descriptive statistics.
minor comments (6)
  1. [Section 3.4 and Section 5] Section 3.4 lists hypotheses H1-H3, but Section 5 refers to 'our hypothesis H4' and 'partially confirming our hypothesis H4'; please define H4 or correct the label.
  2. [Figure 7 caption] The caption states that the asterisk refers to a 'p-value < .5' where p < .05 is clearly intended; as written the caption would include all significant and non-significant results.
  3. [Section 4.6.2 and Section 5] Section 4.6.2 reports significant differences between the Self-taught group and the COM C-XAI, COM A-XAI, and Robot A-XAI groups, with Robot C-XAI as the only non-significant comparison; Section 5 restates this as 'all the others but the COM A-XAI one' and attributes the effect to a COM A-XAI outlier. This inconsistency should be resolved.
  4. [Section 4.5] The sentence 'the former negatively correlated with the percentage of times participants accepted the robot's suggestions (and not the COM's)' is ambiguous; please state explicitly which correlation is reported for which group and report the correlation coefficient.
  5. [Section 4.6] ANOVA results are reported as F(4) without residual degrees of freedom; reporting the full F(df1, df2) notation, e.g., F(4, 50), would aid verification of the analyses.
  6. [General] The paper does not state whether data or analysis scripts are available; a data availability statement would strengthen reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the empirical group comparison is self-contained; the anomaly-as-exploration reading is an acknowledged interpretive risk, not a constructional equivalence.

full rationale

This is an empirical user study rather than a formal derivation, and its central result does not reduce to its inputs. The main claim, that the Self-taught group outperformed assisted groups on the post-experiment test (Section 4.6.2), is supported by ANOVA and Bonferroni post-hoc tests on recorded test answers. No parameter is fitted to those outcomes and then reported as a prediction; the explanation strategies are algorithmic variants of the same decision-tree policy, so comparing them is not tautological. The interpretive step that 'the number of anomalies during training is a good estimator of participants' degree of exploration' (Sections 4.6.2 and 5) is load-bearing but not circular: exploration is not operationally defined as anomaly count, nor are test scores defined in terms of anomalies. The inference is therefore an empirical assumption rather than an identity. The paper itself acknowledges the main associated threat in the limitations paragraph, stating that it is unclear whether the results would hold 'using a fixed number of steps' instead of a fixed time window; this is a validity and mediation concern, not a circularity. The self-citations to prior work by the same group, including the under-review footnote on the assessment task, are used as methodological or contextual provenance and are not invoked to force the empirical outcome. No equation equates an input with an output, and no fitted quantity is renamed as a prediction, so no circular step is present.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The study introduces no new theoretical entities. Its central claims rest on the simulation being a valid learning task, on the AI being expert, and on statistical assumptions that are untested at these sample sizes. The main free parameters are the AI reward design and the classical explanation selection rule, both of which shape the experimental contrast.

free parameters (2)
  • RL reward function weights = not specified
    The reward function for the CQI algorithm is designed by the authors to maximize energy production and avoid damage, determining the AI's optimal policy. The central comparison of explanation strategies depends on the AI being expert; variations in reward weights would change the decision tree and possibly the results.
  • Classical XAI feature selection order = first nodes in DT structure, with a no-repeat heuristic
    Classical explanations select the 'most relevant' features as the first encountered in the DT, a design choice that defines the contrast between C-XAI and A-XAI. It affects the content of explanations and thus participant behavior.
assumptions (3)
  • domain assumption The NPP simulation is a valid and learnable model of nuclear power plant management
    The authors simplified the actual NPP functioning to create a tractable task; the validity of the task as a learning environment is assumed.
  • domain assumption The CQI-trained decision tree is an expert policy for the simulation
    The AI's suggestions are treated as correct or expert in the study; no external validation of optimality is provided.
  • standard math Independent-samples t-tests and ANOVAs are appropriate for the small sample sizes
    Parametric tests are used with n=11-22 per group; normality and homogeneity of variance are not tested.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Let people fail! Exploring the influence of explainable virtual and robotic agents in learning-by-doing tasks." pith.science (2026). https://pith.science/paper/7GPBSEYB

@misc{pith2026241110176,
  author       = {Pith},
  title        = {Pith review of: Let people fail! Exploring the influence of explainable virtual and robotic agents in learning-by-doing tasks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7GPBSEYB}},
  note         = {Machine review of arXiv:2411.10176}
}
read the original abstract

Collaborative decision-making with artificial intelligence (AI) agents presents opportunities and challenges. While human-AI performance often surpasses that of individuals, the impact of such technology on human behavior remains insufficiently understood, primarily when AI agents can provide justifiable explanations for their suggestions. This study compares the effects of classic vs. partner-aware explanations on human behavior and performance during a learning-by-doing task. Three participant groups were involved: one interacting with a computer, another with a humanoid robot, and a third one without assistance. Results indicated that partner-aware explanations influenced participants differently based on the type of artificial agents involved. With the computer, participants enhanced their task completion times. At the same time, those interacting with the humanoid robot were more inclined to follow its suggestions, although they did not reduce their timing. Interestingly, participants autonomously performing the learning-by-doing task demonstrated superior knowledge acquisition than those assisted by explainable AI (XAI). These findings raise profound questions and have significant implications for automated tutoring and human-AI collaboration.

Figures

Figures reproduced from arXiv: 2411.10176 by the authors.

Figure 1
Figure 1. A participant interacts with the humanoid robot iCub during the training phase (Robot group). On the screen is the nuclear power plant application running. 3 Material and Methods This study aims to investigate the informativeness of contrastive explanations (A-XAI) compared to classical causal explanations (C-XAI) during a learning-by-doing task and explore the impact of the explainable agent type on participants’ b… view at source ↗
Figure 2
Figure 2. Experimental design for the NPP experiment. Three macro-groups were considered: COM (with the computer), Self-taught, and Robot. Participants from the COM and Robot groups were divided into two experimental conditions (C-XAI and A-XAI) depending on the explanation strategy adopted by the agent. non-expert users, simple governing rules, the potential for an AI model to learn these rules, and the typical lack of knowl… view at source ↗
Figure 3
Figure 3. Example of DT where the leaf nodes 11 and 2 are the robot’s suggestion and the predicted user’s action, respectively. The classical XAI selects node 1 for the explanations since it is the most unused relevant node. The partner-aware XAI selects node 7 instead because it represents a perfect contrastive explanation for the fact 11 and foil 2. particular, it justifies the agent’s suggestions using the most relevant fe… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Distribution of the participant’s knowledge about the functioning of nuclear power plants before the experiment. We classified them into three levels of knowledge (No, Some, A lot) by coding their open-ended questions to the pre-experiment questionnaire. • H3: particip…
Figure 5
Figure 5. Figure 5: The number of actions participants in the COM group performed during training. The * represents a statistically significant difference (p-value = .039, independent samples t-test). p < .001), with the A-XAI group moving significantly faster than the C-XAI one. Since pa…
Figure 6
Figure 6. Figure 6: Average and std error of participants’ decision times (COM group). Each point of the x-axis represents 5% of the training phase: it has to be read from the left to the right. The plot on the up-left side shows the average decision time of all steps; the one on the up-r…
Figure 7
Figure 7. Figure 7: Explanations’ length measured as the number of words used by the artificial agents. The * refers to a significant difference with p − value < .5, and the *** refers to a strong significant statistical difference with p − value < .001 (independent samples t-test). To co…
Figure 8
Figure 8. Figure 8: Average and std error of participants’ move types during the training phase. The plot on the left refers to the step in which participants asked what questions, while the one on the right refers to the step in which they also asked why questions. It has to be noted tha…
Figure 9
Figure 9. Figure 9: Number of actions performed during the training phase by all the groups. The * refers to strong statistical significance (ANOVA test with Bonferroni correction). asked to report the causes of an indicated action, given precise conditions of the NPP - (ANOVA F(4) = 6.37…
Figure 10
Figure 10. Figure 10: Number of anomalies produced during the training phase by all the groups. The * refers to strong statistical significance (ANOVA test with Bonferroni correction). groups. Although such distributions were comparable, we could note a difference between those referring t…
Figure 11
Figure 11. Figure 11: Average and std error of participants’ decision times during training. Each point of the x-axis represents 5% of the training phase: it has to be read from the left to the right. The plot on the left compares the decision times of the Self-taught group and the COM gro…
Figure 12
Figure 12. Figure 12: Distribution of the participants’ correct answers to the whole test (left) and to the scenarios questions (right). The * refers to p-values between .02 and .001. time was mainly due to the explanation type that participants received and not to an individual difference…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

74 extracted references · 24 canonical work pages

  1. [1]

    Peeking inside the black-box: A survey on explainable artificial intelligence (xai)

    Amina Adadi and Mohammed Berrada. Peeking inside the black-box: A survey on explainable artificial intelligence (xai). IEEE Access, 6: 0 52138--52160, 2018. doi:https://doi.org/10.1109/ACCESS.2018.2870052

    arXiv 2018

  2. [2]

    Highlights: Summarizing agent behavior to people

    Dan Amir and Ofra Amir. Highlights: Summarizing agent behavior to people. In Proceedings of the 17th International Conference on Autonomous Agents and MultiAgent Systems, AAMAS '18, page 1168–1176, Richland, SC, 2018. International Foundation for Autonomous Agents and Multiagent Systems. doi:10.5555/3237383.3237869

    arXiv 2018

  3. [3]

    Explainable agents and robots: Results from a systematic literature review

    Sule Anjomshoae, Amro Najjar, Davide Calvaresi, and Kary Fr\" a mling. Explainable agents and robots: Results from a systematic literature review. In Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems, AAMAS '19, page 1078–1088, Richland, SC, 2019. International Foundation for Autonomous Agents and Multiagent Syst...

    work page arXiv 2019

  4. [4]

    The theory of learning by doing

    Yuichiro Anzai and Herbert A Simon. The theory of learning by doing. Psychological review, 86 0 (2): 0 124, 1979. doi:10.1037/0033-295X.86.2.124. URL https://doi.org/10.1037/0033-295X.86.2.124

    doi:10.1037/0033-295x.86.2.124 1979

  5. [5]

    How cognitive biases affect xai-assisted decision-making: A systematic review

    Astrid Bertrand, Rafik Belloum, James R Eagan, and Winston Maxwell. How cognitive biases affect xai-assisted decision-making: A systematic review. In Proceedings of the 2022 AAAI/ACM conference on AI, ethics, and society, pages 78--91, 2022. doi:10.1145/3514094.3534164

    arXiv 2022

  6. [6]

    Can you trust the black box? the effect of personality traits on trust in ai-enabled user interfaces

    Martin B \"o ckle, Kwaku Yeboah-Antwi, and Iana Kouris. Can you trust the black box? the effect of personality traits on trust in ai-enabled user interfaces. In International Conference on Human-Computer Interaction, pages 3--20. Springer, 2021. doi:10.1007/978-3-030-77772-2\_1

    doi:10.1007/978-3-030-77772-2 2021

  7. [7]

    Zana Bu c inca, Maja Barbara Malaya, and Krzysztof Z. Gajos. To trust or to think: Cognitive forcing functions can reduce overreliance on ai in ai-assisted decision-making. Proc. ACM Hum.-Comput. Interact., 5 0 (CSCW1), apr 2021. doi:10.1145/3449287. URL https://doi.org/10.1145/3449287

    doi:10.1145/3449287 2021

  8. [8]

    Complexity measures and decision tree complexity: a survey

    Harry Buhrman and Ronald de Wolf . Complexity measures and decision tree complexity: a survey. Theoretical Computer Science, 288 0 (1): 0 21--43, 2002. doi:10.1016/S0304-3975(01)00144-X. Complexity and Logic

    doi:10.1016/s0304-3975(01)00144-x 2002

Show all 74 references
  1. [9]

    Plan explanations as model reconciliation: Moving beyond explanation as soliloquy

    Tathagata Chakraborti, Sarath Sreedharan, Yu Zhang, and Subbarao Kambhampati. Plan explanations as model reconciliation: Moving beyond explanation as soliloquy. In Proceedings of the 26th International Joint Conference on Artificial Intelligence, IJCAI'17, page 156–163. AAAI P...

    2017

  2. [11]

    Methods and standards for research on explainable artificial intelligence: lessons from intelligent tutoring systems

    William J Clancey and Robert R Hoffman. Methods and standards for research on explainable artificial intelligence: lessons from intelligent tutoring systems. Applied AI Letters, 2 0 (4): 0 e53, 2021. doi:10.1002/ail2.53

    2021 doi

  3. [12]

    Toward personalized xai: A case study in intelligent tutoring systems

    Cristina Conati, Oswald Barral, Vanessa Putnam, and Lea Rieger. Toward personalized xai: A case study in intelligent tutoring systems. Artificial Intelligence, 298: 0 103503, 2021. doi:10.1016/j.artint.2021.103503

    2021

  4. [13]

    Thinking, fast and slow

    Kahneman Daniel. Thinking, fast and slow. 2017

    2017

  5. [14]

    Wearable reasoner: Towards enhanced human rationality through a wearable device with an explainable ai assistant

    Valdemar Danry, Pat Pataranutaporn, Yaoli Mao, and Pattie Maes. Wearable reasoner: Towards enhanced human rationality through a wearable device with an explainable ai assistant. AHs '20, New York, NY, USA, 2020. Association for Computing Machinery. ISBN 9781450376037. doi:10.1...

    2020

  6. [15]

    An implemented theory of mind to improve human-robot shared plans execution

    Sandra Devin and Rachid Alami. An implemented theory of mind to improve human-robot shared plans execution. In 2016 11th ACM/IEEE International Conference on Human-Robot Interaction (HRI), pages 319--326, 2016. doi:10.1109/HRI.2016.7451768

    2016

  7. [16]

    The human-ai relationship in decision-making: Ai explanation to support people on justifying their decisions

    Juliana Jansen Ferreira and Mateus Monteiro. The human-ai relationship in decision-making: Ai explanation to support people on justifying their decisions. arXiv preprint arXiv:2102.05460, 2021

    2021 arXiv

  8. [17]

    Explainable artificial intelligence for education and training

    Krzysztof Fiok, Farzad V Farahani, Waldemar Karwowski, and Tareq Ahram. Explainable artificial intelligence for education and training. The Journal of Defense Modeling and Simulation, 19 0 (2): 0 133--144, 2022. doi:10.1177/15485129211028651

    2022 doi

  9. [18]

    On cognitive preferences and the plausibility of rule-based models

    Johannes F\" u rnkranz, Tom\' a s Kliegr, and Heiko Paulheim. On cognitive preferences and the plausibility of rule-based models. Mach. Learn., 109 0 (4): 0 853–898, apr 2020. ISSN 0885-6125. doi:10.1007/s10994-019-05856-5. URL https://doi.org/10.1007/s10994-019-05856-5

    2020 doi

  10. [19]

    Considering the context to build theory in hci, hri, and hmc: Explicating differences in processes of communication and socialization with social technologies

    Andrew Gambino and Bingjie Liu. Considering the context to build theory in hci, hri, and hmc: Explicating differences in processes of communication and socialization with social technologies. Human-Machine Communication, 4: 0 111–130, 2022. doi:10.3316/informit.461421232674383

    2022 doi

  11. [20]

    Geiskkovitch, Daniel J

    Denise Y. Geiskkovitch, Daniel J. Rea, Agape Y. Seo, Stela H. Seo, Brittany Postnikoff, and James E. Young. Where should i sit? exploring the impact of seating arrangement in a human-robot collaborative task. In Proceedings of the 8th International Conference on Human-Agent In...

    2020

  12. [21]

    Vera Liao, Yunfeng Zhang, Rachel Bellamy, and Klaus Mueller

    Bhavya Ghai, Q. Vera Liao, Yunfeng Zhang, Rachel Bellamy, and Klaus Mueller. Explainable active learning (xal): Toward ai explanations as interfaces for machine teachers. Proceedings ACM Human-Computer Interaction, 4 0 (CSCW3), jan 2021. doi:10.1145/3432934

    2021 doi

  13. [22]

    Designing ai support for human involvement in ai-assisted decision making: A taxonomy of human-ai interactions from a systematic review

    Catalina Gomez, Sue Min Cho, Chien-Ming Huang, and Mathias Unberath. Designing ai support for human involvement in ai-assisted decision making: A taxonomy of human-ai interactions from a systematic review. arXiv preprint arXiv:2310.19778, 2023

    arXiv

  14. [23]

    Behavior explanation as intention signaling in human-robot teaming

    Ze Gong and Yu Zhang. Behavior explanation as intention signaling in human-robot teaming. In 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), pages 1005--1011, 2018. doi:10.1109/ROMAN.2018.8525675

    2018

  15. [24]

    Counterfactual visual explanations

    Yash Goyal, Ziyan Wu, Jan Ernst, Dhruv Batra, Devi Parikh, and Stefan Lee. Counterfactual visual explanations. In International Conference on Machine Learning, pages 2376--2384. PMLR, 2019

    2019

  16. [25]

    The principles and limits of algorithm-in-the-loop decision making

    Ben Green and Yiling Chen. The principles and limits of algorithm-in-the-loop decision making. Proc. ACM Hum.-Comput. Interact., 3 0 (CSCW), nov 2019. doi:10.1145/3359152. URL https://doi.org/10.1145/3359152

    2019 doi

  17. [26]

    Santos, and Rui Prada

    Mojgan Hashemian, Ana Paiva, Samuel Mascarenhas, Pedro A. Santos, and Rui Prada. The power to persuade: a study of social power in human-robot interaction. In 2019 28th IEEE International Conference on Robot and Human Interactive Communication (RO-MAN), pages 1--8, 2019. doi:1...

    2019

  18. [27]

    Metrics for explainable ai: Challenges and prospects

    Robert R Hoffman, Shane T Mueller, Gary Klein, and Jordan Litman. Metrics for explainable ai: Challenges and prospects. arXiv preprint arXiv:1812.04608, 2018

    2018 arXiv

  19. [28]

    Causability and explainability of artificial intelligence in medicine

    Andreas Holzinger, Georg Langs, Helmut Denk, Kurt Zatloukal, and Heimo M \"u ller. Causability and explainability of artificial intelligence in medicine. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 9 0 (4): 0 e1312, 2019. doi:10.1002/widm.1312

    2019 doi

  20. [29]

    u ller. Measuring the quality of explanations: the system causability scale (scs). KI-K \

    Andreas Holzinger, Andr \'e Carrington, and Heimo M \"u ller. Measuring the quality of explanations: the system causability scale (scs). KI-K \"u nstliche Intelligenz , 34 0 (2): 0 193--198, 2020. doi:10.1007/s13218-020-00636-z

    2020 doi

  21. [30]

    Will algorithms blind people? the effect of explainable ai and decision-makers’ experience on ai-supported decision-making in government

    Marijn Janssen, Martijn Hartog, Ricardo Matheus, Aaron Yi Ding, and George Kuk. Will algorithms blind people? the effect of explainable ai and decision-makers’ experience on ai-supported decision-making in government. Social Science Computer Review, 40 0 (2): 0 478--493, 2022....

    2022 doi

  22. [31]

    Representativeness revisited: Attribute substitution in intuitive judgment

    Daniel Kahneman, Shane Frederick, et al. Representativeness revisited: Attribute substitution in intuitive judgment. Heuristics and biases: The psychology of intuitive judgment, 49 0 (49-81): 0 74, 2002. doi:10.1017/CBO9780511808098.004. URL https://doi.org/10.1017/CBO97805118...

    2002 doi

  23. [32]

    Personalised self-explanation by robots: The role of goals versus beliefs in robot-action explanation for children and adults

    Frank Kaptein, Joost Broekens, Koen Hindriks, and Mark Neerincx. Personalised self-explanation by robots: The role of goals versus beliefs in robot-action explanation for children and adults. In 2017 26th IEEE International Symposium on Robot and Human Interactive Communicatio...

    2017

  24. [33]

    Explainable artificial intelligence in education

    Hassan Khosravi, Simon Buckingham Shum, Guanliang Chen, Cristina Conati, Yi-Shan Tsai, Judy Kay, Simon Knight, Roberto Martinez-Maldonado, Shazia Sadiq, and Dragan Gašević. Explainable artificial intelligence in education. Computers and Education: Artificial Intelligence, 3: 0...

    2022

  25. [34]

    Too much, too little, or just right? ways explanations impact end users' mental models

    Todd Kulesza, Simone Stumpf, Margaret Burnett, Sherry Yang, Irwin Kwan, and Weng-Keen Wong. Too much, too little, or just right? ways explanations impact end users' mental models. In 2013 IEEE Symposium on Visual Languages and Human Centric Computing, pages 3--10, 2013. doi:10...

    2013

  26. [35]

    An evaluation of the human-interpretability of explanation

    Isaac Lage, Emily Chen, Jeffrey He, Menaka Narayanan, Been Kim, Sam Gershman, and Finale Doshi-Velez. An evaluation of the human-interpretability of explanation. arXiv preprint arXiv:1902.00006, 2019

    1902 arXiv

  27. [36]

    Towards a science of human-ai decision making: a survey of empirical studies

    Vivian Lai, Chacha Chen, Q Vera Liao, Alison Smith-Renner, and Chenhao Tan. Towards a science of human-ai decision making: a survey of empirical studies. arXiv preprint arXiv:2112.11471, 2021

    2021 arXiv

  28. [37]

    Lim, Anind K

    Brian Y. Lim, Anind K. Dey, and Daniel Avrahami. Why and why not explanations improve the intelligibility of context-aware intelligent systems. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI '09, page 2119–2128, New York, NY, USA, 2009. Asso...

    2009

  29. [38]

    Who should i trust: Ai or myself? leveraging human and ai correctness likelihood to promote appropriate trust in ai-assisted decision-making

    Shuai Ma, Ying Lei, Xinru Wang, Chengbo Zheng, Chuhan Shi, Ming Yin, and Xiaojuan Ma. Who should i trust: Ai or myself? leveraging human and ai correctness likelihood to promote appropriate trust in ai-assisted decision-making. In Proceedings of the 2023 CHI Conference on Huma...

    2023

  30. [39]

    a mling. Explainable agents for less bias in human-agent decision making. In Davide Calvaresi, Amro Najjar, Michael Winikoff, and Kary Fr \

    Avleen Malhi, Samanta Knapic, and Kary Fr \"a mling. Explainable agents for less bias in human-agent decision making. In Davide Calvaresi, Amro Najjar, Michael Winikoff, and Kary Fr \"a mling, editors, Explainable, Transparent Autonomous Agents and Multi-Agent Systems, pages 1...

    2020 doi

  31. [40]

    knowing me, knowing you

    Millecamp Martijn, Cristina Conati, and Katrien Verbert. “knowing me, knowing you”: personalized explanations for a music recommender system. User Modeling and User-Adapted Interaction, 32 0 (1): 0 215--252, 2022. doi:10.1007/s11257-021-09304-9

    2022 doi

  32. [41]

    A user-centred framework for explainable artificial intelligence in human-robot interaction

    Marco Matarese, Francesco Rea, and Alessandra Sciutti. A user-centred framework for explainable artificial intelligence in human-robot interaction. arXiv preprint arXiv:2109.12912, 2021

    2021 arXiv

  33. [42]

    Ex(plainable) machina: how social-implicit xai affects complex human-robot teaming tasks

    Marco Matarese, Francesca Cocchella, Francesco Rea, and Alessandra Sciutti. Ex(plainable) machina: how social-implicit xai affects complex human-robot teaming tasks. In 2023 IEEE International Conference on Robotics and Automation (ICRA), pages 11986--11993, 2023 a . doi:10.11...

    2023

  34. [43]

    Natural born explainees: how users’ personality traits shape the human-robot interaction with explainable robots

    Marco Matarese, Francesca Cocchella, Francesco Rea, and Alessandra Sciutti. Natural born explainees: how users’ personality traits shape the human-robot interaction with explainable robots. In 2023 32nd IEEE International Conference on Robot and Human Interactive Communication...

    2023

  35. [44]

    To explain or not to explain: The effects of personal characteristics when explaining music recommendations

    Martijn Millecamp, Nyi Nyi Htun, Cristina Conati, and Katrien Verbert. To explain or not to explain: The effects of personal characteristics when explaining music recommendations. In Proceedings of the 24th International Conference on Intelligent User Interfaces, IUI '19, page...

    2019

  36. [45]

    What's in a user? towards personalising transparency for music recommender interfaces

    Martijn Millecamp, Nyi Nyi Htun, Cristina Conati, and Katrien Verbert. What's in a user? towards personalising transparency for music recommender interfaces. In Proceedings of the 28th ACM Conference on User Modeling, Adaptation and Personalization, UMAP '20, page 173–182, New...

    2020

  37. [46]

    Explainable recommendations and calibrated trust: Two systematic user errors

    Mohammad Naiseh, Deniz Cemiloglu, Dena Al Thani, Nan Jiang, and Raian Ali. Explainable recommendations and calibrated trust: Two systematic user errors. Computer, 54 0 (10): 0 28--37, 2021. doi:10.1109/MC.2021.3076131

    2021

  38. [47]

    o rg Schl \

    Meike Nauta, Jan Trienes, Shreyasi Pathak, Elisa Nguyen, Michelle Peters, Yasmin Schmitt, J \"o rg Schl \"o tterer, Maurice van Keulen, and Christin Seifert. From anecdotal evidence to quantitative evaluation methods: A systematic review on evaluating explainable ai. arXiv pre...

    2022 arXiv

  39. [48]

    Mollik Nayyar, Zachary Zoloty, Ciera McFarland, and Alan R. Wagner. Exploring the effect of explanations during robot-guided emergency evacuation. In Social Robotics, pages 13--22. Springer International Publishing, 2020. doi:10.1007/978-3-030-62056-1_2

    2020 doi

  40. [49]

    Conservative q-improvement: Reinforcement learning for an interpretable decision-tree policy

    Aaron M Roth, Nicholay Topin, Pooyan Jamshidi, and Manuela Veloso. Conservative q-improvement: Reinforcement learning for an interpretable decision-tree policy. arXiv preprint arXiv:1907.01180, 2019

    1907 arXiv

  41. [50]

    How robots influence humans: A survey of nonverbal communication in social human--robot interaction

    Shane Saunderson and Goldie Nejat. How robots influence humans: A survey of nonverbal communication in social human--robot interaction. International Journal of Social Robotics, 11: 0 575--608, 2019. doi:10.1007/s12369-019-00523-0

    2019 doi

  42. [51]

    Investigating strategies for robot persuasion in social human–robot interaction

    Shane Saunderson and Goldie Nejat. Investigating strategies for robot persuasion in social human–robot interaction. IEEE Transactions on Cybernetics, 52 0 (1): 0 641--653, 2022. doi:10.1109/TCYB.2020.2987463

    2022

  43. [52]

    Learning by doing

    Roger C Schank, Tamara R Berman, and Kimberli A Macpherson. Learning by doing. In Instructional-design theories and models, pages 161--181. Routledge, 2013

    2013

  44. [53]

    Should i follow ai-based advice? measuring appropriate reliance in human-ai decision-making

    Max Schemmer, Patrick Hemmer, Niklas K \"u hl, Carina Benz, and Gerhard Satzger. Should i follow ai-based advice? measuring appropriate reliance in human-ai decision-making. arXiv preprint arXiv:2204.06916, 2022 a

    2022 arXiv

  45. [54]

    u hl, and Michael V \

    Max Schemmer, Patrick Hemmer, Maximilian Nitsche, Niklas K \"u hl, and Michael V \"o ssing. A meta-analysis on the utility of explainable artificial intelligence in human-ai decision-making. arXiv preprint arXiv:2205.05126, 2022 b

    2022 arXiv

  46. [55]

    Explainable robotics in human-robot interactions

    Rossitza Setchi, Maryam Banitalebi Dehkordi, and Juwairiya Siraj Khan. Explainable robotics in human-robot interactions. Procedia Computer Science, 176: 0 3057--3066, 2020. doi:10.1016/j.procs.2020.09.198

    2020 doi

  47. [56]

    The human side of xai: Bridging the gap between ai and non-expert audiences

    Beatriz Severes, Carolina Carreira, Ana Beatriz Vieira, Eduardo Gomes, Jo\ a o Tiago Apar\' cio, and In\^ e s Pereira. The human side of xai: Bridging the gap between ai and non-expert audiences. In Proceedings of the 41st ACM International Conference on Design of Communicatio...

    2023

  48. [57]

    new” and “old

    Auste Simkute, Aditi Surana, Ewa Luger, Michael Evans, and Rhianne Jones. Xai for learning: Narrowing down the digital divide between “new” and “old” experts. In Adjunct Proceedings of the 2022 Nordic Human-Computer Interaction Conference, NordiCHI '22, New York, NY, USA, 2022...

    2022

  49. [58]

    How to quantify the degree of explainability: Experiments and practical implications

    Francesco Sovrano and Fabio Vitali. How to quantify the degree of explainability: Experiments and practical implications. In 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), pages 1--9. IEEE, 2022. doi:10.1109/FUZZ-IEEE55066.2022.9882574

    2022

  50. [59]

    Selecting optimal split-functions for large datasets

    Kilian Stoffel and Laura E Raileanu. Selecting optimal split-functions for large datasets. In Research and Development in Intelligent Systems XVII, pages 62--72. Springer, 2001. doi:10.1007/978-1-4471-0269-4\_5

    2001 doi

  51. [60]

    Towards explainable multi-objective probabilistic planning

    Roykrong Sukkerd, Reid Simmons, and David Garlan. Towards explainable multi-objective probabilistic planning. In Proceedings of the 4th International Workshop on Software Engineering for Smart Cyber-Physical Systems, SEsCPS '18, page 19–25, New York, NY, USA, 2018. Association...

    2018

  52. [61]

    Evaluating the effectiveness of explanations for recommender systems

    Nava Tintarev and Judith Masthoff. Evaluating the effectiveness of explanations for recommender systems. User Modeling and User-Adapted Interaction, 22 0 (4): 0 399--439, 2012. doi:10.1007/s11257-011-9117-5

    2012 doi

  53. [63]

    Evaluating xai: A comparison of rule-based and example-based explanations

    Jasper van der Waa , Elisabeth Nieuwburg, Anita Cremers, and Mark Neerincx. Evaluating xai: A comparison of rule-based and example-based explanations. Artificial Intelligence, 291, 2021. doi:10.1016/j.artint.2020.103404

    2021

  54. [64]

    Decision tree extraction using trained neural network

    Nikola Vasilev, Zheni Mincheva, and Ventsislav Nikolov. Decision tree extraction using trained neural network. In SMARTGREENS, pages 194--200, 2020

    2020

  55. [65]

    The effects of explanations on automation bias

    Mor Vered, Tali Livni, Piers Douglas Lionel Howe, Tim Miller, and Liz Sonenberg. The effects of explanations on automation bias. Artificial Intelligence, 322: 0 103952, 2023. ISSN 0004-3702. doi:https://doi.org/10.1016/j.artint.2023.103952. URL https://www.sciencedirect.com/sc...

    2023

  56. [66]

    Governing algorithmic decisions: The role of decision importance and governance on perceived legitimacy of algorithmic decisions

    Ari Waldman and Kirsten Martin. Governing algorithmic decisions: The role of decision importance and governance on perceived legitimacy of algorithmic decisions. Big Data & Society, 9 0 (1): 0 20539517221100449, 2022. doi:10.1177/20539517221100449. URL https://doi.org/10.1177/...

    2022 doi

  57. [67]

    Explainable embodied agents through social cues: A review

    Sebastian Wallk\" o tter, Silvia Tulli, Ginevra Castellano, Ana Paiva, and Mohamed Chetouani. Explainable embodied agents through social cues: A review. J. Hum.-Robot Interact., 10 0 (3), jul 2021. doi:10.1145/3457188. URL https://doi.org/10.1145/3457188

    2021 doi

  58. [68]

    Investigating explainable human-robot interaction with augmented reality

    Chao Wang and Anna Belardinelli. Investigating explainable human-robot interaction with augmented reality. In 5th International Workshop on Virtual, Augmented, and Mixed Reality for HRI, 2022. URL https://openreview.net/forum?id=S2zeAWt4hk5

    2022

  59. [69]

    Danding Wang, Qian Yang, Ashraf Abdul, and Brian Y. Lim. Designing theory-driven user-centric explainable ai. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, CHI '19, page 1–15, New York, NY, USA, 2019. Association for Computing Machinery. ISBN...

    2019

  60. [70]

    Scout: Self-aware discriminant counterfactual explanations

    Pei Wang and Nuno Vasconcelos. Scout: Self-aware discriminant counterfactual explanations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8981--8990, 2020. doi:10.1109/CVPR42600.2020.00900

    2020

  61. [71]

    Are explanations helpful? a comparative study of the effects of explanations in ai-assisted decision-making

    Xinru Wang and Ming Yin. Are explanations helpful? a comparative study of the effects of explanations in ai-assisted decision-making. In 26th International Conference on Intelligent User Interfaces, IUI '21, page 318–328, New York, NY, USA, 2021. Association for Computing Mach...

    2021

  62. [72]

    Effects of explanations in ai-assisted decision making: Principles and comparisons

    Xinru Wang and Ming Yin. Effects of explanations in ai-assisted decision making: Principles and comparisons. ACM Transactions on Interactive Intelligent Systems (TiiS), 2022. doi:10.1145/3519266

    2022 doi

  63. [73]

    Wason and J.ST.B.T

    P.C. Wason and J.ST.B.T. Evans. Dual processes in reasoning? Cognition, 3 0 (2): 0 141--154, 1974. ISSN 0010-0277. doi:https://doi.org/10.1016/0010-0277(74)90017-1. URL https://www.sciencedirect.com/science/article/pii/0010027774900171

    1974

  64. [74]

    Towards human-centred explainable ai: A systematic literature review

    Oyindamola Williams. Towards human-centred explainable ai: A systematic literature review. 2021

    2021

  65. [75]

    Learning decision trees with reinforcement learning

    Zheng Xiong, Wenpeng Zhang, and Wenwu Zhu. Learning decision trees with reinforcement learning. In NIPS Workshop on Meta-Learning, 2017

    2017

  66. [76]

    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 12, 2026 · model on record in the stance chip above.