August 2023
Volume 23, Issue 9
Open Access
Vision Sciences Society Annual Meeting Abstract  |   August 2023
Enhancing Multi-Category Perceptual Learning Using Signal Detection Theory Concepts in Dermatologic Cancer Screening
Author Affiliations & Notes
  • Philip Kellman
    University of California, Los Angeles
  • Sally Krasne
    University of California, Los Angeles
  • Everett Mettler
    University of California, Los Angeles
  • Timothy Burke
    University of California, Los Angeles
  • Christine Massey
    University of California, Los Angeles
  • Footnotes
    Acknowledgements  National Institutes of Health​/National Cancer Institute​ grant R01CA236791
Journal of Vision August 2023, Vol.23, 5565. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Philip Kellman, Sally Krasne, Everett Mettler, Timothy Burke, Christine Massey; Enhancing Multi-Category Perceptual Learning Using Signal Detection Theory Concepts in Dermatologic Cancer Screening. Journal of Vision 2023;23(9):5565.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Combining perceptual learning techniques with adaptive learning algorithms has been shown to accelerate the development of expertise in medical and STEM learning domains. Virtually all adaptive learning systems have relied on simple accuracy data. In multi-category perceptual classifications, however, accuracy in responding to exemplars of a category may be misleading if the learner has high sensitivity and poor specificity (i.e., correct responses may occur from either competence or bias). We investigated whether adaptive perceptual learning in dermatologic screening can be enhanced by incorporating signal detection theory (SDT) methods that separate true sensitivity from criterion. SDT-style concepts were used to alter spacing, such that recurrence depended both on accuracy for that category and false alarms to other categories. SDT concepts were also used to define mastery (category retirement) criteria: instead of retirement based on accuracy and response times, SDT retirement used a running d’ estimate calculated from a recent window of trials based on hit and false alarm rates. We used a Dermatologic Screening PALM (perceptual adaptive learning module) to train classification of 10 cancerous and non-cancerous skin lesion types. Four adaptive conditions varied either the type of adaptive spacing (standard vs. SDT) or retirement criteria (standard vs. SDT). A control condition presented standard didactic instruction on dermatologic screening in video form, including images to illustrate and explain structures commonly used to classify malignant lesions and benign look-alikes (e.g., ABCDE Criteria, reticular networks, pigmentation patterns, vascular patterns). Results: All adaptive conditions outperformed the control in both learning efficiency and fluency (large effect sizes). Use of SDT retirement criteria produced greater learning efficiency at both immediate and delayed posttests (medium effect sizes). SDT spacing and standard adaptive spacing did not differ. Conclusion: Some SDT enhancements to adaptive perceptual learning procedures may improve learning efficiency and fluency.  


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.