Pith. sign in

REVIEW 3 major objections 5 minor 93 references

What Clinicians Need: Designing, Developing and Evaluating an AI-Based Decision Support System for Autism Assessment

T0 review · 3 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read A clinician-centered AI system for autism screening changed the decision paths of clinicians in an evaluation study and earned their support as a screening aid and training tool.

desk verdict A solid, honest HCI study whose real contribution is the decision-path analysis, not the AI model; the label-provenance gap is a genuine but fixable weakness. read the letter →

arxiv 2607.22005 v1 pith:JZPLOWCW submitted 2026-07-24 cs.HC

Ulrike Schäfer , William Saakyan , Matthias Norden , Fabrizio Kuruc , Peter Sörries , Isabel Dziobek , Claudia Müller-Birn , Hanna Drimalla This is my paper
classification cs.HC
keywords clinicaldecisionsupportsystemautismspectrumconditionadultscreeningnonverbalbehavioranalysishuman-AIcollaborationmentalmodelspathsclinicianevaluation
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 tries to show that an AI decision support system for adult autism screening can be built from clinicians' stated needs and that it measurably changes the way clinicians reason about a case. From interviews with seven clinicians, the authors identified five challenges—high demand, subjective assessments, information overload, workflow fit, and expectation setting—and turned them into design strategies. They then built SIT-CARE, which visualizes gaze, facial expressions, and voice from a standardized interaction and adds an AI recommendation on whether further autism diagnostics are warranted. In an evaluation with seven new clinicians assessing two patient cases at three decision points each, the authors found four distinct decision paths, with clinicians sometimes revising, consolidating, or holding their judgments after seeing the data or the AI output. The paper concludes that such a system can support screening, deepen diagnostic assessment, and help less experienced clinicians learn to read nonverbal behavior.

What carries the argument

The load-bearing object is SIT-CARE itself, a web-based clinical decision support system with two modes. The data-based assessment mode shows three clinically relevant modalities—gaze behavior, intensity and variability of facial expressions, and voice pitch variability—using line plots, violin plots, and a bi-dimensional gaze histogram, with gender-specific reference groups and prototypical ASC and Non-ASC cases. The model-based assessment mode provides a recommendation from a late-fusion binary classifier trained on 1,140 nonverbal features from 325 adults, presented with a predicted probability, a contextualized confusion matrix, and warnings about limitations. The mechanism that carries

What would settle it

A blinded prospective study with, say, 40 independently diagnosed adults: if SIT-CARE's AI recommendation matches verified diagnoses no better than existing self-report questionnaires, or if clinicians change decisions just as often with a control interface, the claim that the system changes and improves decision-making fails.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that SIT-CARE, a clinical decision support system designed with and for clinicians, does not simply add an AI recommendation but restructures the decision-making process. Its data-based assessment mode supports earlier stages of judgment—forming hypotheses and gathering evidence—by presenting visualizations of gaze, facial expressions, and voice alongside reference groups and prototypical examples. Its model-based assessment mode gives a binary AI recommendation with confidence information and stated limitations. When seven clinicians assessed two real patient cases at three decision points—after the video, after exploring the data, and afte

Load-bearing premise

The conclusions rest on the assumption that the 325 training labels and the two evaluation cases (one ASC, one Non-ASC) are diagnostically correct, yet the paper describes no independent verification of those diagnoses.

Editorial extensions

If this is right

  • If SIT-CARE is used in screening, it could complement or replace self-report questionnaires of disputed utility in adults.
  • The system supports earlier decision-making stages, such as hypothesis generation and data gathering, not just the final recommendation.
  • Because the standardized interaction can be self-administered in minutes and the system is web-based, the approach could scale to telehealth and help relieve referral bottlenecks.
  • Less experienced clinicians can use the data-based mode with prototypical examples and reference bands to build diagnostic sensitivity for nonverbal behavior.
  • Variation in clinicians' mental models and decision paths implies that AI support should be paired with training, onboarding, and transparent communication of accuracy and limitations.

Reading between the lines

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

  • Editorial inference: If the four decision-path types prove stable in a larger sample, the taxonomy could be used to adapt the interface to a clinician's reasoning style rather than offering one fixed presentation.
  • Editorial inference: Because one group of clinicians consistently disagreed with the AI while another ignored it, the next natural test is whether explanations calibrated to an individual's mental model reduce both under-reliance and over-reliance.
  • Editorial inference: The inclusion of gender-specific reference groups may counteract the known underdiagnosis of autistic women, but that is a testable consequence the paper does not itself claim.
  • Editorial inference: The reported 74% cross-validated accuracy and the two-case evaluation leave open how the tool behaves under differential diagnosis; a natural next experiment would pit SIT-CARE against cases with depression, ADHD, or eating disorders.
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

3 major / 5 minor

Summary. The paper describes the design, development, and evaluation of SIT-CARE, an AI-based clinical decision support system (CDSS) for adult autism spectrum condition (ASC) screening. A formative study with seven clinicians elicited five challenges and corresponding design strategies; these informed a web-based system with two modes: a data-based assessment showing visualizations of gaze, facial expressions, and voice, and a model-based assessment providing an AI recommendation with confidence information. The evaluation recruited seven new clinicians who assessed two patient cases (one ASC, one Non-ASC) at three decision points (video only, after data-based assessment, after model-based assessment), yielding 14 decision paths. The authors identify four groups of decision paths, characterize clinicians' mental models of the AI, and report that clinicians perceived SIT-CARE as useful for screening and as a learning opportunity. The central claim is that SIT-CARE demonstrated potential to improve initial diagnostic assessments and to empower less experienced clinicians.

Significance. If the claims are substantiated, the paper makes a worthwhile contribution to human-centered AI in clinical decision support. It offers one of the few end-to-end examples of a CDSS for adult ASC screening that is explicitly grounded in formative interviews, and it provides a qualitative decision-path taxonomy that goes beyond simple accuracy metrics. The computational transparency in Appendix D—including fixed random seeds, the anonymization noise level, and the distance-to-median score formula—is a concrete strength. The authors also openly acknowledge limitations in Section 7 (small sample, need for quantitative follow-up, cultural generalizability). However, the significance is tempered by the fact that the evaluation is qualitative with seven clinicians and two deliberately easy cases, and by the absence of any stated gold standard for the ASC/Non-ASC labels that underlie both model training and the interpretation of the decision paths.

major comments (3)
  1. [§4.2 / §5.1] The paper never states how the ASC/Non-ASC labels were determined for the 325 training participants or for the two evaluation cases. Section 4.2 lists inclusion criteria (age, IQ, fluency, pharmacotherapy, comorbidity) but not the diagnostic procedure (e.g., ADOS/ADI-R, consensus clinical diagnosis, or self-report). Section 5.1 describes the two cases only as 'chosen by experienced clinicians based on their diagnostic assessments' and 'considered clear cases.' Because the model-based assessment is the final decision point in every path in Figure 3, the 'AI confirmed' and 'changed to align with AI' paths assume the AI recommendation is meaningful. The reported 74% LOOCV accuracy and the statement that the model 'classified correctly' the two cases cannot be evaluated without this provenance. Given the paper's own citation that ADOS has only ~65% sensitivity in adults (§2.1), the label sou
  2. [Abstract / §5 / §7] The conclusion that SIT-CARE 'demonstrated potential in improving initial diagnostic assessments' is not directly supported by the study design. The evaluation measures decision changes and self-reported usefulness, not diagnostic accuracy, clinician learning, or any quantitative outcome related to assessment quality. Seven clinicians and two deliberately easy cases cannot establish improvement in initial assessments; at best they show that decision paths differ and that clinicians find the system helpful. The limitations in Section 7 acknowledge the need for large-scale quantitative work, but the abstract and conclusion should be more precise: the evidence supports potential value in supporting decision-making and training, not demonstrated improvement in diagnostic accuracy.
  3. [§5.1 / Fig. 3] The two evaluation cases were deliberately selected as clear cases and were both correctly classified by the AI model. This design creates a ceiling effect: all 'AI confirmed' paths involve a correct AI recommendation, and there are no trials in which the AI is wrong or ambiguous. As a result, the mental-model findings cannot speak to over-reliance or under-reliance when the AI errs, which is exactly the risk discussed in Section 6.1. The authors should either include cases with incorrect or uncertain AI recommendations, or explicitly restrict the claims about reliance to the explored setting rather than drawing general conclusions about over- and under-reliance.
minor comments (5)
  1. [§5.3.1] 'In seven of the fourteen decisions to be made' should almost certainly read 'seven of the fourteen decision paths'; as written, the number does not match the 42 individual decisions made in the study.
  2. [§3.1 / §5.1] IRB approval numbers and institutions are omitted with 'will be provided after acceptance.' Please include this information or state why it is withheld.
  3. [Fig. 3] The decision-path figure is dense and hard to parse. Consider adding group labels, per-group path counts, and a clearer legend for the arrows and decision labels.
  4. [Appendix D.2] The distance-to-median score formula is ambiguous: the notation 'median_j(...)' and the denominator structure are not defined precisely. Please define all indices and operations so the computation is reproducible.
  5. [§4.2] The sentence 'The model was then applied to classify two patient cases ... which the model classified correctly' should specify the exact model version and feature set used for these two cases, and ideally link to the evaluation descriptions in Section 5.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the model is evaluated with participant-wise LOOCV on held-out participants, and the decision-path findings come from newly collected clinician interviews.

full rationale

The paper's central claims are empirical and qualitative rather than derived from a fitted parameter. The AI model is trained on 325 participants and evaluated with participant-based Leave-One-Out Cross-Validation (Section 4.2), which is a standard non-circular protocol; the two evaluation cases are explicitly reported as not part of the training dataset. The decision-path and mental-model findings are grounded in newly recorded think-aloud and interview data from seven newly recruited clinicians (Section 5), not in the model's training labels. The use of the SIT and prior accuracy results is supported by citations to the authors' earlier work [24, 69], but these are background references for the choice of paradigm rather than load-bearing derivations of the present results, and the paper independently retrains and reevaluates its own model. The unresolved question of how ASC/Non-ASC labels were established for the training and evaluation cases is a data-provenance and external-validity concern, not a circularity by construction: no equation or fitted parameter is renamed as a prediction. Therefore no circular step can be exhibited from the manuscript text.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central claim depends on both a qualitative design process and an ML model. The ML model and reference norms come from the authors' own prior SIT dataset, and the free parameters listed above are fitting/choice decisions not independently validated. No new physical or conceptual entities (e.g., particles, forces) are introduced.

free parameters (5)
  • Fixation density histogram thresholds = ±0.5σ and ±1.5σ relative to Non-ASC reference group
    Five discrete color levels chosen by hand in Appendix D.1; no justification for these standard-deviation cutoffs.
  • Representative frame sampling seed = random_state=42, 5,500 frames
    Fixed seed and frame count chosen arbitrarily in Appendix D.1 to illustrate prototypical gaze distributions.
  • Anonymization noise level = 15% Gaussian noise
    Added to prototypical facial-expression traces in Section 4.2; chosen for privacy, not clinically validated.
  • Distance-to-median score formula = Score combining total_abs_diff and per-phase variance deviations
    Constructed in Appendix D.2 to select representative individuals; no external validation of this selection criterion.
  • ML model hyperparameters = Not fully reported (XGBoost, logistic regression with polynomial degree 2)
    The late-fusion model in Section 4.2 is fitted to the 325-participant dataset; hyperparameters and classification threshold are not disclosed.
assumptions (6)
  • domain assumption The SIT standardized interaction elicits clinically relevant nonverbal behavior for adult ASC assessment.
    Taken from prior work [24]; the paper does not independently validate this assumption beyond citing the SIT literature.
  • domain assumption Ground-truth ASC/Non-ASC labels in the 325-participant training set are correct.
    Section 4.2 lists inclusion criteria but does not describe independent diagnostic verification for the training labels.
  • domain assumption The two evaluation patient cases are correctly diagnosed and 'clear cases'.
    Section 5.1 states they were chosen by experienced clinicians, but no diagnostic gold-standard details are provided.
  • domain assumption OpenFace and openSMILE features validly measure gaze, facial expressions, and voice parameters relevant to ASC.
    Tooling assumptions in Section 4.2; the paper relies on these tools' outputs without independent validation in this context.
  • domain assumption Purposive samples of seven clinicians per study are informative about the broader target user population.
    The authors acknowledge the small sample and call for larger quantitative studies in Section 7.
  • domain assumption Training data from one country generalizes to other cultural contexts.
    Explicitly flagged as a limitation in Section 7: 'nonverbal behavior is strongly influenced by culture.'

how reviews work

0 comments
Cite this review

Pith. "Pith review of What Clinicians Need: Designing, Developing and Evaluating an AI-Based Decision Support System for Autism Assessment." pith.science (2026). https://pith.science/paper/JZPLOWCW

@misc{pith2026260722005,
  author       = {Pith},
  title        = {Pith review of: What Clinicians Need: Designing, Developing and Evaluating an AI-Based Decision Support System for Autism Assessment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JZPLOWCW}},
  note         = {Machine review of arXiv:2607.22005}
}
read the original abstract

AI methods promise to support autism spectrum condition (ASC) diagnostics in adults, a complex and time-consuming process, that is characterized by a shortage of specialized clinicians. To date, clinicians' needs and their interaction with such AI-based support remain underexplored. Our work aims to develop and evaluate an AI-based clinical decision support system (CDSS) for ASC assessment, and to investigate how it impacts clinicians' decision-making. By interviewing clinicians of varying experience levels, we identified five challenges and derived design strategies. Based on that, we developed SIT-CARE, a CDSS, which provides AI-based recommendations and data visualizations of clinically relevant nonverbal behavior. Through an evaluation study with newly recruited clinicians, we found that SIT-CARE led to different decision paths in regard to the ASC assessment, which are reflected in clinicians' mental models and decision changes. Overall, SIT-CARE demonstrated potential in improving initial diagnostic assessments, supporting in-depth diagnosis and empowering less experienced clinicians.

Figures

Figures reproduced from arXiv: 2607.22005 by the authors.

Figure 2
Figure 2. View of model-based assessment of SIT-CARE including the AI recommendation for a hypothetical patient and warnings about limitations. Left: Sidebar with collapsible additional information about the AI model and significance. behavior. In SIT-CARE, these extracted features are used to generate visualizations and numerical summaries of gaze behavior, facial expressions, and voice parameters within the data-based asses… view at source ↗
Figure 6
Figure 6. Visualization of the chose voice data variability of pitch intensity per phase, of a hypothetical Non-ASC patient. [PITH_FULL_IMAGE:figures/full_fig_p027_6.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

93 extracted references · 36 canonical work pages

  1. [1]

    Valerie Carl, , and Oliver Hinz

    Benjamin Abdel-Karim, Nicolas Pfeuffer, K. Valerie Carl, , and Oliver Hinz. 2023. How AI-Based Systems Can Induce Reflections: The Case of AI-Augmented Diagnostic Work.MIS Quarterly47, 4 (Dec. 2023), 1395–1424. doi:10.25300/MISQ/2022/16773

    work page doi:10.25300/misq/2022/16773 2023

  2. [2]

    Bennett, Kori Inkpen, Jaime Teevan, Ruth Kikin-Gil, and Eric Horvitz

    Saleema Amershi, Dan Weld, Mihaela Vorvoreanu, Adam Fourney, Besmira Nushi, Penny Collisson, Jina Suh, Shamsi Iqbal, Paul N. Bennett, Kori Inkpen, Jaime Teevan, Ruth Kikin-Gil, and Eric Horvitz. 2019. Guidelines for Human-AI Interaction. InProceedings of the 2019 CHI Conference on Human Factors in Computing Systems. ACM, Glasgow Scotland Uk, 1–13. doi:10....

    arXiv 2019

  3. [3]

    Chittaranjan Andrade. 2021. The Inconvenient Truth About Convenience and Purposive Samples.Indian Journal of Psychological Medicine43, 1 (Jan. 2021), 86–88. doi:10.1177/0253717620977000

    work page doi:10.1177/0253717620977000 2021

  4. [4]

    K. L. Ashwood, N. Gillan, J. Horder, H. Hayward, E. Woodhouse, F. S. McEwen, J. Findon, H. Eklund, D. Spain, C. E. Wilson, T. Cadman, S. Young, V. Stoencheva, C. M. Murphy, D. Robertson, T. Charman, P. Bolton, K. Glaser, P. Asherson, E. Simonoff, and D. G. Murphy. 2016. Predicting the Diagnosis of Autism in Adults Using the Autism-Spectrum Quotient (AQ) Q...

    work page doi:10.1017/s0033291716001082 2016

  5. [5]

    2013/2013.Diagnostic and Statistical Manual of Mental Disorders : DSM-5™.(5th edition

    American Psychiatric Association. 2013/2013.Diagnostic and Statistical Manual of Mental Disorders : DSM-5™.(5th edition. ed.). American Psychiatric Publishing, a division of American Psychiatric Association, Washington, DC

    2013

  6. [6]

    If I Had All the Time in the World

    Anne Kathrine Petersen Bach, Trine Munch Nørgaard, Jens Christian Brok, and Niels Van Berkel. 2023. “If I Had All the Time in the World”: Ophthalmologists’ Perceptions of Anchoring Bias Mitigation in Clinical AI Support. InProceedings of the 2023 CHI Conference on Human Factors in Computing Systems. ACM, Hamburg Germany, 1–14. doi:10.1145/3544548.3581513

    arXiv 2023

  7. [7]

    Tadas Baltrusaitis, Amir Zadeh, Yao Chong Lim, and Louis-Philippe Morency. 2018. OpenFace 2.0: Facial Behavior Analysis Toolkit. In2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018). IEEE, Xi’an, China, 59–66. doi:10.1109/FG.2018.00019

    arXiv 2018

  8. [8]

    Lasecki, Daniel S

    Gagan Bansal, Besmira Nushi, Ece Kamar, Walter S. Lasecki, Daniel S. Weld, and Eric Horvitz. 2019. Beyond Accuracy: The Role of Mental Models in Human-AI Team Performance.Proceedings of the AAAI Conference on Human Computation and Crowdsourcing7 (Oct. 2019), 2–11. doi:10.1609/hcomp.v7i1.5285

    doi:10.1609/hcomp.v7i1.5285 2019

Show all 93 references
  1. [9]

    Gagan Bansal, Besmira Nushi, Ece Kamar, Daniel S Weld, Walter S Lasecki, and Eric Horvitz. 2019. Updates in human-ai teams: Understanding and addressing the performance/compatibility tradeoff, In Proceedings of the AAAI Conference on Artificial Intelligence.Proceedings of the ...

    2019 doi

  2. [10]

    Sarah Bargiela, Robyn Steward, and William Mandy. 2016. The experiences of late-diagnosed women with autism spectrum conditions: An investigation of the female autism phenotype.Journal of autism and developmental disorders46, 10 (2016), 3281–3294

    2016

  3. [11]

    Simon Baron-Cohen, Sally Wheelwright, Richard Skinner, Joanne Martin, and Emma Clubley. 2001. The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/High-Functioning Autism, Males and Females, Scientists and Mathematicians.Journal of Autism and Developmental Disord...

    2001 doi

  4. [12]

    Alejandro Barredo Arrieta, Natalia Díaz-Rodríguez, Javier Del Ser, Adrien Bennetot, Siham Tabik, Alberto Barbado, Salvador Garcia, Sergio Gil-Lopez, Daniel Molina, Richard Benjamins, Raja Chatila, and Francisco Herrera. 2020. Explainable Artificial Intelligence (XAI): Concepts...

    2020 doi

  5. [13]

    Vardoulakis

    Emma Beede, Elizabeth Baylor, Fred Hersch, Anna Iurchenko, Lauren Wilcox, Paisan Ruamviboonsuk, and Laura M. Vardoulakis. 2020. A Human- Centered Evaluation of a Deep Learning System Deployed in Clinics for the Detection of Diabetic Retinopathy. InProceedings of the 2020 CHI C...

    2020

  6. [14]

    Kellen Briot, Adrien Pizano, Manuel Bouvard, and Anouck Amestoy. 2021. New Technologies as Promising Tools for Assessing Facial Emotion Expressions Impairments in ASD: A Systematic Review.Frontiers in Psychiatry12 (2021), 634756. doi:10.3389/fpsyt.2021.634756

    2021

  7. [15]

    Ángel Alexander Cabrera, Adam Perer, and Jason I. Hong. 2023. Improving Human-AI Collaboration With Descriptions of AI Behavior.Proceedings of the ACM on Human-Computer Interaction7, CSCW1 (April 2023), 1–21. doi:10.1145/3579612

    2023 doi

  8. [16]

    H. Pinar Caglar-Nazali, Freya Corfield, Valentina Cardi, Suman Ambwani, Jenni Leppanen, Olaolu Olabintan, Stephanie Deriziotis, Alexandra Hadjimichalis, Pasquale Scognamiglio, Ertimiss Eshkevari, Nadia Micali, and Janet Treasure. 2014. A systematic review and meta-analysis of ...

    2014 doi

  9. [17]

    Hello AI

    Carrie J. Cai, Samantha Winter, David Steiner, Lauren Wilcox, and Michael Terry. 2019. "Hello AI": Uncovering the Onboarding Needs of Medical Practitioners for Human-AI Collaborative Decision-Making.Proceedings of the ACM on Human-Computer Interaction3, CSCW (Nov. 2019), 1–24....

    2019 doi

  10. [18]

    Shiye Cao, Anqi Liu, and Chien-Ming Huang. 2024. Designing for Appropriate Reliance: The Roles of AI Uncertainty Presentation, Initial User Decision, and User Demographics in AI-Assisted Decision-Making.Proceedings of the ACM on Human-Computer Interaction8, CSCW1 (April 2024),...

    2024 doi

  11. [19]

    Tara Capel and Margot Brereton. 2023. What is Human-Centered about Human-Centered AI? A Map of the Research Landscape. InProceedings of the 2023 CHI Conference on Human Factors in Computing Systems. ACM, Hamburg Germany, 1–23. doi:10.1145/3544548.3580959

    2023

  12. [20]

    Lawan, Zurki Ibrahim, Abdullahi Dahiru, Sadiya Tahir, Usama Ishaq Abdulrazak, and Adamu Hussaini

    Nadire Cavus, Abdulmalik A. Lawan, Zurki Ibrahim, Abdullahi Dahiru, Sadiya Tahir, Usama Ishaq Abdulrazak, and Adamu Hussaini. 2021. A Systematic Literature Review on the Application of Machine-Learning Models in Behavioral Assessment of Autism Spectrum Disorder.Journal of Pers...

    2021 doi

  13. [21]

    Maxwell Harper, and Haiyi Zhu

    Hao-Fei Cheng, Ruotong Wang, Zheng Zhang, Fiona O’Connell, Terrance Gray, F. Maxwell Harper, and Haiyi Zhu. 2019. Explaining Decision-Making Algorithms through UI: Strategies to Help Non-Expert Stakeholders. InProceedings of the 2019 CHI Conference on Human Factors in Computin...

    2019

  14. [22]

    Conner, Ryan D

    Caitlin M. Conner, Ryan D. Cramer, and John J. McGonigle. 2019. Examining the Diagnostic Validity of Autism Measures Among Adults in an Outpatient Clinic Sample.Autism in Adulthood: Challenges and Management1, 1 (March 2019), 60–68. doi:10.1089/aut.2018.0023

    2019

  15. [23]

    Conor James Davidson, Alwyn Kam, Frances Needham, and Alison Jane Stansfield. 2015. No exclusions – developing an autism diagnostic service for adults irrespective of intellectual ability.Advances in Autism1, 2 (Oct. 2015), 66–78. doi:10.1108/AIA-08-2015-0010

    2015 doi

  16. [24]

    Hanna Drimalla, Tobias Scheffer, Niels Landwehr, Irina Baskow, Stefan Roepke, Behnoush Behnia, and Isabel Dziobek. 2020. Towards the automatic detection of social biomarkers in autism spectrum disorder: introducing the simulated interaction task (SIT).npj Digital Medicine3, 1 ...

    2020 doi

  17. [25]

    Florian Eyben, Martin Wöllmer, and Björn Schuller. 2010. Opensmile: The Munich Versatile and Fast Open-Source Audio Feature Extractor. In Proceedings of the 18th ACM International Conference on Multimedia(2010-10-25)(MM ’10). Association for Computing Machinery, New York, NY, ...

    2010

  18. [26]

    Laura Fusar-Poli, Natascia Brondino, Pierluigi Politi, and Eugenio Aguglia. 2022. Missed diagnoses and misdiagnoses of adults with autism spectrum disorder.European Archives of Psychiatry and Clinical Neuroscience272, 2 (March 2022), 187–198. doi:10.1007/s00406-020-01189-w

    2022 doi

  19. [27]

    Is voice a marker for Autism spectrum disorder? A systematic review and meta-analysis

    Riccardo Fusaroli, Anna Lambrechts, Dan Bang, Dermot M. Bowler, and Sebastian B. Gaigg. 2017. “Is voice a marker for Autism spectrum disorder? A systematic review and meta-analysis”.Autism Research10, 3 (March 2017), 384–407. doi:10.1002/aur.1678 Publisher: Wiley

    2017 doi

  20. [28]

    Maria Gendron, Carlos Crivelli, and Lisa Feldman Barrett. 2018. Universality reconsidered: Diversity in making meaning of facial expressions. Current directions in psychological science27, 4 (2018), 211–219

    2018

  21. [29]

    Millen, Murray Campbell, Sadhana Kumaravel, and Wei Zhang

    Katy Ilonka Gero, Zahra Ashktorab, Casey Dugan, Qian Pan, James Johnson, Werner Geyer, Maria Ruiz, Sarah Miller, David R. Millen, Murray Campbell, Sadhana Kumaravel, and Wei Zhang. 2020. Mental Models of AI Agents in a Cooperative Game Setting. InProceedings of the 2020 CHI Co...

    2020

  22. [30]

    2009.The Elements of Statistical Learning: Data Mining, Inference, and Prediction(2nd ed.)

    Trevor Hastie, Robert Tibshirani, and Jerome Friedman. 2009.The Elements of Statistical Learning: Data Mining, Inference, and Prediction(2nd ed.). Springer, New York. https://link.springer.com/book/10.1007/978-0-387-84858-7 See Chapter 7 for cross-validation methods, including...

    2009 doi

  23. [31]

    Patrick Hemmer, Max Schemmer, Michael Vössing, and Niklas Kühl. 2021. Human-AI Complementarity in Hybrid Intelligence Systems: A Structured Literature Review. https://www.researchgate.net/publication/352882174_Human-AI_Complementarity_in_Hybrid_Intelligence_Systems_A_ Structur...

    2021

  24. [32]

    Hoffman, Shane T

    Robert R. Hoffman, Shane T. Mueller, Gary Klein, and Jordan Litman. 2023. Measures for explainable AI: Explanation goodness, user satisfaction, mental models, curiosity, trust, and human-AI performance.Frontiers in Computer Science5 (Feb. 2023), 1096257. doi:10.3389/fcomp.2023.1096257

    2023

  25. [33]

    Kenneth Holstein, Maria De-Arteaga, Lakshmi Tumati, and Yanghuidi Cheng. 2023. Toward Supporting Perceptual Complementarity in Human-AI Collaboration via Reflection on Unobservables.Proceedings of the ACM on Human-Computer Interaction7, CSCW1 (April 2023), 1–20. doi:10.1145/ m...

    2023

  26. [34]

    Laura Hull, KV Petrides, and William Mandy. 2020. The female autism phenotype and camouflaging: A narrative review.Review journal of autism and developmental disorders7, 4 (2020), 306–317

    2020

  27. [35]

    Pradier, Barbara Lam, Andrew C

    Maia Jacobs, Jeffrey He, Melanie F. Pradier, Barbara Lam, Andrew C. Ahn, Thomas H. McCoy, Roy H. Perlis, Finale Doshi-Velez, and Krzysztof Z. Gajos. 2021. Designing AI for Trust and Collaboration in Time-Constrained Medical Decisions: A Sociotechnical Lens. InProceedings of th...

    2021

  28. [36]

    Sarah L Jones, Maria Johnson, Bronwen Alty, and Marios Adamou. 2021. The Effectiveness of RAADS-R as a Screening Tool for Adult ASD Populations.Autism Research and Treatment2021, 1 (2021), 9974791. doi:10.1155/2021/9974791

    2021 doi

  29. [37]

    Shahad Sabbar Joudar, A. S. Albahri, Rula A. Hamid, Idrees A. Zahid, M. E. Alqaysi, O. S. Albahri, and A. H. Alamoodi. 2023. Artificial intelligence- based approaches for improving the diagnosis, triage, and prioritization of autism spectrum disorder: a systematic review of cu...

    2023 doi

  30. [38]

    Markelle Kelly, Aakriti Kumar, Padhraic Smyth, and Mark Steyvers. 2023. Capturing Humans’ Mental Models of AI: An Item Response Theory Approach. In2023 ACM Conference on Fairness, Accountability, and Transparency. ACM, Chicago IL USA, 1723–1734. doi:10.1145/3593013.3594111

    2023

  31. [39]

    Paul, Daniel P

    Na Yeon Kim, Junfeng He, Qianying Wu, Na Dai, Kai Kohlhoff, Jasmin Turner, Lynn K. Paul, Daniel P. Kennedy, Ralph Adolphs, and Vidhya Navalpakkam. 2024. Smartphone-Based Gaze Estimation for in-Home Autism Research.Autism Research17, 6 (2024), 1140–1148. doi:10.1002/aur.3140

    2024 doi

  32. [40]

    Anne-Kathrin Kleine, Eesha Kokje, Pia Hummelsberger, Eva Lermer, Insa Schaffernak, and Susanne Gaube. 2025. AI-enabled clinical decision support tools for mental healthcare: A product review.Artificial Intelligence in Medicine160 (Feb. 2025), 103052. doi:10.1016/j.artmed.2024.103052

    2025

  33. [41]

    Kolbinger, Gregory P

    Fiona R. Kolbinger, Gregory P. Veldhuizen, Jiefu Zhu, Daniel Truhn, and Jakob Nikolas Kather. 2024. Reporting Guidelines in Medical Artificial Intelligence: A Systematic Review and Meta-Analysis.Communications Medicine4, 1 (April 2024), 71. doi:10.1038/s43856-024-00492-0

    2024 doi

  34. [42]

    Todd Kulesza, Simone Stumpf, Margaret Burnett, and Irwin Kwan. 2012. Tell Me More?: The Effects of Mental Model Soundness on Personalizing an Intelligent Agent. InProceedings of the SIGCHI Conference on Human Factors in Computing Systems. ACM, Austin Texas USA, 1–10. doi:10.11...

    2012

  35. [43]

    Anil Kumar and Umesh Chandra Jaiswal. 2025. Predicting Autism Spectrum Disorder in Adults Through Facial Image Analysis: A Multi-CNN with BiLSTM Model.SN Comput. Sci.6, 3 (March 2025). doi:10.1007/s42979-025-03783-y

    2025 doi

  36. [44]

    2022.Empowering Humans in Human-AI Decision Making

    Vivian Lai. 2022.Empowering Humans in Human-AI Decision Making. Dissertation. University of Colorado, Colorado. https: //www.proquest.com/openview/a9d79c3540c48b3853cebb6bc5b35d12/1?cbl=18750&diss=y&pq-origsite=gscholar&parentSessionId= 1IjMGURB9SKpdC4mrFueGWhE4HmqZj9IF08FQfKGIlo%3D

    2022

  37. [45]

    Vivian Lai, Samuel Carton, Rajat Bhatnagar, Q Vera Liao, Yunfeng Zhang, and Chenhao Tan. 2022. Human-AI Collaboration via Conditional Delegation: A Case Study of Content Moderation. InCHI Conference on Human Factors in Computing Systems. ACM, New Orleans LA USA, 1–18. doi:10.1...

    2022

  38. [46]

    Vivian Lai, Chacha Chen, Alison Smith-Renner, Q Vera Liao, and Chenhao Tan. 2023. Towards a Science of Human-AI Decision Making: An Overview of Design Space in Empirical Human-Subject Studies. In2023 ACM Conference on Fairness, Accountability, and Transparency. ACM, Chicago IL...

    2023

  39. [47]

    Anish Lakkapragada, Aaron Kline, Onur Cezmi Mutlu, Kelley Paskov, Brianna Chrisman, Nathaniel Stockham, Peter Washington, and Dennis Paul Wall. 2022. The Classification of Abnormal Hand Movement to Aid in Autism Detection: Machine Learning Study.JMIR Biomedical Engineering7, 1...

    2022 doi

  40. [49]

    Blanke, Ulrike Suenkel, and Isabel Dziobek

    Silke Lipinski, Elisabeth S. Blanke, Ulrike Suenkel, and Isabel Dziobek. 2019. Outpatient Psychotherapy for Adults with High-Functioning Autism Spectrum Condition: Utilization, Treatment Satisfaction, and Preferred Modifications.Journal of Autism and Developmental Disorders49,...

    2019 doi

  41. [50]

    Silke Lipinski, Katharina Boegl, Elisabeth S Blanke, Ulrike Suenkel, and Isabel Dziobek. 2022. A blind spot in mental healthcare? Psychotherapists lack education and expertise for the support of adults on the autism spectrum.Autism26, 6 (Aug. 2022), 1509–1521. doi:10.1177/1362...

    2022 doi

  42. [51]

    Robyn Longhurst. 2003. Semi-structured interviews and focus groups.Key methods in geography3, 2 (2003), 143–156

    2003

  43. [52]

    Catherine Lord, Susan Risi, Linda Lambrecht, Edwin H Cook Jr, Bennett L Leventhal, Pamela C DiLavore, Andrew Pickles, and Michael Rutter. 2000. The Autism Diagnostic Observation Schedule—Generic: A standard measure of social and communication deficits associated with the spect...

    2000

  44. [53]

    Catherine Lord, Michael Rutter, and Ann Le Couteur. 1994. Autism Diagnostic Interview—Revised: A Revised Version of a Diagnostic Interview for Caregivers of Individuals with Possible Pervasive Developmental Disorders.Journal of Autism and Developmental Disorders24, 5 (1994), 6...

    1994 doi

  45. [54]

    2014.Qualitative content analysis: theoretical foundation, basic procedures and software solution

    Philipp Mayring. 2014.Qualitative content analysis: theoretical foundation, basic procedures and software solution. Klagenfurt, Austria

    2014

  46. [55]

    Goldie A McQuaid, Nancy Raitano Lee, and Gregory L Wallace. 2022. Camouflaging in autism spectrum disorder: Examining the roles of sex, gender identity, and diagnostic timing.Autism26, 2 (Feb. 2022), 552–559. doi:10.1177/13623613211042131

    2022 doi

  47. [56]

    Alessio Maria Monteleone, Giammarco Cascino, Valeria Ruzzi, Niccolò Marafioti, Luigi Marone, Roberta Croce Nanni, and Alfonso Troisi. 2022. Non-verbal social communication in individuals with eating disorders: an ethological analysis in experimental setting.Eating and Weight D...

    2022 doi

  48. [57]

    Stefan Morana, Silvia Schacht, Ansgar Scherp, and Alexander Maedche. 2017. A review of the nature and effects of guidance design features. Decision Support Systems97 (May 2017), 31–42. doi:10.1016/j.dss.2017.03.003

    2017 doi

  49. [58]

    Katelyn Morrison, Philipp Spitzer, Violet Turri, Michelle Feng, Niklas Kühl, and Adam Perer. 2024. The Impact of Imperfect XAI on Human-AI Decision-Making.Proceedings of the ACM on Human-Computer Interaction8, CSCW1 (April 2024), 1–39. doi:10.1145/3641022

    2024 doi

  50. [59]

    Walsh, and Karen M

    Ranjani Narayanan, Sarah E. Walsh, and Karen M. Feigh. 2023. Development of Mental Models in Decision-Making Tasks.Proceedings of the Human Factors and Ergonomics Society Annual Meeting67, 1 (Sept. 2023), 767–773. doi:10.1177/21695067231192195

    2023 doi

  51. [60]

    1988.The Psychology of Everyday Things.Basic books

    Donald A Norman. 1988.The Psychology of Everyday Things.Basic books

    1988

  52. [61]

    Donald A Norman. 2014. Some Observations on Mental Models. InMental Models. Psychology Press, London, 15–22

    2014

  53. [62]

    2004.ICD-10 : international statistical classification of diseases and related health problems : tenth revision

    World Health Organization. 2004.ICD-10 : international statistical classification of diseases and related health problems : tenth revision. World Health Organization

    2004

  54. [63]

    World Health Organization. 2022. ICD-11: International classification of diseases (11th revision). https://icd.who.int/ Accessed: 2023-09-10

    2022

  55. [64]

    Marc Pinski, Martin Adam, and Alexander Benlian. 2023. AI Knowledge: Improving AI Delegation through Human Enablement. InProceedings of the 2023 CHI Conference on Human Factors in Computing Systems. ACM, Hamburg Germany, 1–17. doi:10.1145/3544548.3580794

    2023

  56. [65]

    Antonin Poché, Lucas Hervier, and Mohamed-Chafik Bakkay. 2023. Natural Example-Based Explainability: A Survey. InExplainable Artificial Intelligence, Luca Longo (Ed.). Springer Nature Switzerland, Cham, 24–47

    2023

  57. [66]

    font-variant:small-caps

    Jacqueline A. Riddiford, Peter G. Enticott, Alex Lavale, and Caroline Gurvich. 2022. Gaze and social functioning associations in autism spectrum disorder: A systematic review and <span style="font-variant:small-caps;">meta-analysis</span>.Autism Research15, 8 (Aug. 2022), 1380...

    2022 doi

  58. [67]

    Ritvo, Donald Guthrie, Max J

    Riva Ariella Ritvo, Edward R. Ritvo, Donald Guthrie, Max J. Ritvo, Demetra H. Hufnagel, William McMahon, Bruce Tonge, David Mataix-Cols, Amita Jassi, Tony Attwood, and Johann Eloff. 2011. The Ritvo Autism Asperger Diagnostic Scale-Revised (RAADS-R): A Scale to Assist the Diagn...

    2011 doi

  59. [68]

    When Two Wrongs Don’t Make a Right

    Emely Rosbach, Jonas Ammeling, Sebastian Krügel, Angelika Kießig, Alexis Fritz, Jonathan Ganz, Chloé Puget, Taryn Donovan, Andrea Klang, Maximilian C. Köller, Pompei Bolfa, Marco Tecilla, Daniela Denk, Matti Kiupel, Georgios Paraschou, Mun Keong Kok, Alexander F. H. Haake, Ron...

    2025

  60. [69]

    William Saakyan, Matthias Norden, Lola Herrmann, Simon Kirsch, Muyu Lin, Simon Guendelman, Isabel Dziobek, and Hanna Drimalla. 2023. On Scalable and Interpretable Autism Detection from Social Interaction Behavior. In2023 11th International Conference on Affective Computing and...

    2023

  61. [70]

    Max Schemmer, Niklas Kuehl, Carina Benz, Andrea Bartos, and Gerhard Satzger. 2023. Appropriate Reliance on AI Advice: Conceptualization and the Effect of Explanations. InProceedings of the 28th International Conference on Intelligent User Interfaces. ACM, Sydney NSW Australia,...

    2023

  62. [71]

    Kelly A Shaw. 2025. Prevalence and early identification of autism spectrum disorder among children aged 4 and 8 years—Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022.MMWR. Surveillance Summaries74 (2025)

    2025

  63. [72]

    Hong Shen, Haojian Jin, Ángel Alexander Cabrera, Adam Perer, Haiyi Zhu, and Jason I. Hong. 2020. Designing Alternative Representations of Confusion Matrices to Support Non-Expert Public Understanding of Algorithm Performance.Proceedings of the ACM on Human-Computer Interaction...

    2020 doi

  64. [73]

    Roberta Simeoli, Angelo Rega, Mariangela Cerasuolo, Raffaele Nappo, and Davide Marocco. 2024. Using Machine Learning for Motion Analysis to Early Detect Autism Spectrum Disorder: A Systematic Review.Review Journal of Autism and Developmental Disorders(March 2024). doi:10.1007/...

    2024

  65. [74]

    Da-Yea Song, So Yoon Kim, Guiyoung Bong, Jong Myeong Kim, and Hee Jeong Yoo. 2019. The Use of Artificial Intelligence in Screening and Diagnosis of Autism Spectrum Disorder: A Literature Review.Journal of the Korean Academy of Child and Adolescent Psychiatry30, 4 (Oct. 2019), ...

    2019 doi

  66. [75]

    Sox, Michael C

    Harold C. Sox, Michael C. Higgins, Douglas K. Owens, and Gillian Sanders Schmidler. 2024.Medical decision making(third edition ed.). John Wiley & Sons, Inc, Hoboken, NJ. doi:10.1002/9781119627876

    2024 doi

  67. [76]

    Aaron Springer and Steve Whittaker. 2020. Progressive Disclosure: When, Why, and How Do Users Want Algorithmic Transparency Information? ACM Transactions on Interactive Intelligent Systems10, 4 (Dec. 2020), 1–32. doi:10.1145/3374218

    2020 doi

  68. [77]

    Syriopoulou-Delli

    Christine K. Syriopoulou-Delli. 2025. Advances in Autism Spectrum Disorder (ASD) Diagnostics: From Theoretical Frameworks to AI-Driven Innovations.Electronics14, 5 (Jan. 2025), 951. doi:10.3390/electronics14050951

    2025 doi

  69. [78]

    Fadi Thabtah. 2019. An Accessible and Efficient Autism Screening Method for Behavioural Data and Predictive Analyses.Health Informatics Journal 25, 4 (Dec. 2019), 1739–1755. doi:10.1177/1460458218796636

    2019 doi

  70. [79]

    Peter Todd and Izak Benbasat. 1987. Process Tracing Methods in Decision Support Systems Research: Exploring the Black Box.MIS Quarterly11, 4 (Dec. 1987), 493. doi:10.2307/248979 manuscript What Clinicians Need 25

    1987 doi

  71. [80]

    Trevisan, Maureen Hoskyn, and Elina Birmingham

    Dominic A. Trevisan, Maureen Hoskyn, and Elina Birmingham. 2018. Facial Expression Production in Autism: A Meta-Analysis.Autism Research 11, 12 (Dec. 2018), 1586–1601. doi:10.1002/aur.2037 Publisher: Wiley

    2018 doi

  72. [81]

    Jennifer Uekermann, M Kraemer, Mona Abdel-Hamid, Benno Graf Schimmelmann, Johannes Hebebrand, Irene Daum, Jens Wiltfang, and Bernhard Kis. 2010. Social cognition in attention-deficit hyperactivity disorder (ADHD).Neuroscience & biobehavioral reviews34, 5 (2010), 734–743

    2010

  73. [82]

    Michelle Vaccaro, Abdullah Almaatouq, and Thomas Malone. 2024. When combinations of humans and AI are useful: A systematic review and meta-analysis.Nature Human Behaviour8, 12 (Oct. 2024), 2293–2303. doi:10.1038/s41562-024-02024-1

    2024 doi

  74. [83]

    Vitalari

    Nicholas P. Vitalari. 1985. Knowledge as a Basis for Expertise in Systems Analysis: An Empirical Study.MIS Quarterly9, 3 (Sept. 1985), 221. doi:10.2307/248950

    1985 doi

  75. [84]

    Vogeley, J

    K. Vogeley, J. C. Kirchner, A. Gawronski, L. Tebartz Van Elst, and I. Dziobek. 2013. Toward the development of a supported employment program for individuals with high-functioning autism in Germany.European Archives of Psychiatry and Clinical Neuroscience263, S2 (Nov. 2013), 1...

    2013 doi

  76. [85]

    Fred R Volkmar, Marc Woodbury-Smith, S Macari, RA Oien, J McPartland, and D Stavropoulos. 2022. Diagnostic issues and complexities in autism and related conditions.Differential Diagnosis of Autism Spectrum Disorder(2022), 1

    2022

  77. [86]

    Brilliant AI Doctor

    Dakuo Wang, Liuping Wang, Zhan Zhang, Ding Wang, Haiyi Zhu, Yvonne Gao, Xiangmin Fan, and Feng Tian. 2021. “Brilliant AI Doctor” in Rural Clinics: Challenges in AI-Powered Clinical Decision Support System Deployment. InProceedings of the 2021 CHI Conference on Human Factors in...

    2021

  78. [87]

    Danding Wang, Qian Yang, Ashraf Abdul, and Brian Y. Lim. 2019. Designing Theory-Driven User-Centric Explainable AI. InProceedings of the 2019 CHI Conference on Human Factors in Computing Systems. ACM, Glasgow Scotland Uk, 1–15. doi:10.1145/3290605.3300831

    2019

  79. [88]

    Liuping Wang, Zhan Zhang, Dakuo Wang, Weidan Cao, Xiaomu Zhou, Ping Zhang, Jianxing Liu, Xiangmin Fan, and Feng Tian. 2023. Human- centered design and evaluation of AI-empowered clinical decision support systems: a systematic review.Frontiers in Computer Science5 (June 2023), ...

    2023

  80. [89]

    Sara Wolf, Tobias Grundgeiger, Raphael Zähringer, Lora Shishkova, Franzisca Maas, Christina Dilling, and Oliver Happel. 2025. How a Clinical Decision Support System Changed the Diagnosis Process: Insights from an Experimental Mixed-Method Study in a Full-Scale Anesthesiology S...

    2025 doi

  81. [90]

    Barry Wright, Penny Spikins, and Hannah Pearson. 2020. Should Autism Spectrum Conditions Be Characterised in a More Positive Way in Our Modern World?Medicina56, 5 (May 2020), 233. doi:10.3390/medicina56050233

    2020 doi

  82. [91]

    Siyi Wu, Weidan Cao, Shihan Fu, Bingsheng Yao, Ziqi Yang, Changchang Yin, Varun Mishra, Daniel Addison, Ping Zhang, and Dakuo Wang. 2025. CardioAI: A Multimodal AI-based System to Support Symptom Monitoring and Risk Prediction of Cancer Treatment-Induced Cardiotoxicity. In Pro...

    2025

  83. [92]

    Qian Yang, Aaron Steinfeld, and John Zimmerman. 2019. Unremarkable AI: Fitting Intelligent Decision Support into Critical, Clinical Decision- Making Processes. InProceedings of the 2019 CHI Conference on Human Factors in Computing Systems. ACM, Glasgow Scotland Uk, 1–11. doi:1...

    2019

  84. [93]

    Padilla, Jeffrey Caterino, Ping Zhang, and Dakuo Wang

    Shao Zhang, Jianing Yu, Xuhai Xu, Changchang Yin, Yuxuan Lu, Bingsheng Yao, Melanie Tory, Lace M. Padilla, Jeffrey Caterino, Ping Zhang, and Dakuo Wang. 2024. Rethinking Human-AI Collaboration in Complex Medical Decision Making: A Case Study in Sepsis Diagnosis. InProceedings ...

    2024

  85. [94]

    Yujie Zhu, Mingyuan Zhang, Cong Fang, Le Fang, Meichen Liu, Yonghao Long, Kun-Pyo Lee, Lie Zhang, and Stephen Jia Wang. 2025. AI Doctor for ASD: Physician Perceptions and Adoption Challenges in Autism Clinical Practice.Proc. ACM Hum.-Comput. Interact.9, 2 (May 2025), CSCW027:1...

    2025 doi

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.