Papers
Topics
Authors
Recent
AI Research Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 78 tok/s
Gemini 2.5 Pro 42 tok/s Pro
GPT-5 Medium 23 tok/s Pro
GPT-5 High 20 tok/s Pro
GPT-4o 103 tok/s Pro
Kimi K2 188 tok/s Pro
GPT OSS 120B 454 tok/s Pro
Claude Sonnet 4 38 tok/s Pro
2000 character limit reached

Can AI Mitigate Human Perceptual Biases? A Pilot Study (2311.00706v1)

Published 10 Oct 2023 in cs.HC and cs.AI

Abstract: We present results from a pilot experiment to measure if machine recommendations can debias human perceptual biases in visualization tasks. We specifically studied the pull-down'' effect, i.e., people underestimate the average position of lines, for the task of estimating the ensemble average of data points in line charts. These line charts can show for example temperature or precipitation in 12 months. Six participants estimated ensemble averages with or without an AI assistant. The assistant, when available, responded at three different speeds to assemble the conditions of a human collaborator who may delay his or her responses. Our pilot study showed that participants were faster with AI assistance in ensemble tasks, compared to the baseline without AI assistance. Althoughpull-down'' biases were reduced, the effect of AI assistance was not statistically significant. Also, delaying AI responses had no significant impact on human decision accuracy. We discuss the implications of these preliminary results for subsequent studies.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (24)
  1. Visual validation versus visual estimation: A study on the average value in scatterplots. IEEE Trans Vis Comput Graph, 2024.
  2. How to evaluate data visualizations across different levels of understanding. In IEEE Workshop on Evaluation and Beyond-Methodological Approaches to Visualization (BELIV), pp. 19–28, 2020. doi: 10 . 1109/BELIV51497 . 2020 . 00010
  3. Measuring the effects of scalar and spherical colormaps on ensembles of DMRI tubes. IEEE Trans Vis Comput Graph, 26(9):2818–2833, 2019. doi: 10 . 1109/TVCG . 2019 . 2898438
  4. W. S. Cleveland and R. McGill. Graphical perception: Theory, experimentation, and application to the development of graphical methods. Journal of the American Statistical Association, 79(387):531–554, 1984. doi: 10 . 2307/2288400
  5. Comparing averages in time series data. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 1095–1104, 2012. doi: 10 . 1145/2207676 . 2208556
  6. Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 248–255, 2009. doi: 10 . 1109/CVPR . 2009 . 5206848
  7. J. J. Ferreira and M. Monteiro. The human-AI relationship in decision-making: AI explanation to support people on justifying their decisions. Workshop on Transparency And Explanations In Smart Systems (TEXSS) held in conjunction with ACM Intelligent User Interfaces (IUI), 2021. doi: 10 . 48550/arXiv . 2102 . 05460
  8. Arguing machines: Human supervision of black box ai systems that make life-critical decisions. In Proc. CVPR, pp. 0–0, 2019.
  9. C. Frydman and I. Krajbich. Using response times to infer others’ private information: an application to information cascades. Management Science, 68(4):2970–2986, 2022. doi: 10 . 1287/mnsc . 2021 . 3994
  10. Do as ai say: susceptibility in deployment of clinical decision-aids. NPJ digital medicine, 4(1):31, 2021. doi: 10 . 1038/s41746-021-00385-9
  11. Evaluating ‘graphical perception’with cnns. IEEE Trans Vis Comput Graph, 25(1):641–650, 2018. doi: 10 . 1109/TVCG . 2018 . 2865138
  12. Long-term priors influence visual perception through recruitment of long-range feedback. Nat Commun, 12(1):6288, 2021. doi: 10 . 1038/s41467-021-26544-w
  13. The worst of both worlds: A comparative analysis of errors in learning from data in psychology and machine learning. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, pp. 335–348, 2022. doi: 10 . 1145/3514094 . 3534196
  14. CNN prefer large training data span; coupling span and diversity improves accuracy and robustness. IEEE Trans on Vis Comput Graph (under submission), 2023.
  15. Visualizing multivalued data from 2d incompressible flows using concepts from painting. In Proceedings Visualization, pp. 333–540. IEEE, 1999. doi: 10 . 1109/VISUAL . 1999 . 809905
  16. Rethinking explainability as a dialogue: A practitioner’s perspective. arXiv preprint arXiv:2202.01875, 2022. doi: 10 . 48550/arXiv . 2202 . 01875
  17. Assessing the impact of automated suggestions on decision making: Domain experts mediate model errors but take less initiative. In Proc. ACM. (CHI), pp. 1–13, 2021. doi: 10 . 1145/3411764 . 3445522
  18. Illustrative interactive stipple rendering. IEEE Trans Vis Comput Graph, 9(2):127–138, 2003. doi: 10 . 1109/TVCG . 2003 . 1196001
  19. K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition. ICLR, 2015.
  20. Counterfactual explanations can be manipulated. Advances in neural information processing systems (NIPS), 34:62–75, 2021.
  21. A. Tversky and D. Kahneman. Judgment under uncertainty: Heuristics and biases: Biases in judgments reveal some heuristics of thinking under uncertainty. science, 185(4157):1124–1131, 1974. doi: 10 . 1126/science . 185 . 4157 . 1124
  22. D. Whitney and A. Yamanashi Leib. Ensemble perception. Annual Review of Psychology, 69:105–129, 2018. doi: 10 . 1146/annurev-psych-010416-044232
  23. Biased average position estimates in line and bar graphs: Underestimation, overestimation, and perceptual pull. IEEE Trans Vis Comput Graph, 26(1):301–310, 2019. doi: 10 . 1109/TVCG . 2019 . 2934400
  24. The curse of knowledge in visual data communication. IEEE Trans Vis Comput Graph, 26(10):3051–3062, 2019. doi: 10 . 1109/TVCG . 2019 . 2917689

Summary

We haven't generated a summary for this paper yet.

Lightbulb On Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

Don't miss out on important new AI/ML research

See which papers are being discussed right now on X, Reddit, and more:

“Emergent Mind helps me see which AI papers have caught fire online.”

Philip

Philip

Creator, AI Explained on YouTube