Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Time, space and feature similarity determine attractive and repulsive serial biases in trustworthiness impressions.

BMC biology·2026
Same author

Adaptive optical correction for in vivo two-photon fluorescence microscopy with neural fields.

Nature methods·2026
Same author

Continuity fields enhance visual perception through positive serial dependence.

Nature reviews psychology·2026
Same author

Integration of affective cues in context-rich and dynamic scenes varies across individuals.

Nature communications·2025
Same author

Slow change blindness from serial dependence.

bioRxiv : the preprint server for biology·2025
Same author

Predictors of childhood vaccination uptake and timeliness: a cross-sectional study in a diverse urban UK population.

The British journal of general practice : the journal of the Royal College of General Practitioners·2025

Related Experiment Video

Updated: Jul 24, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.9K

Serial dependence in perception across naturalistic generative adversarial network-generated mammogram.

Zhihang Ren1, Teresa Canas-Bajo1, Cristina Ghirardo2

  • 1University of California, Berkeley, Vision Science Graduate Group, Berkeley, California, United States.

Journal of Medical Imaging (Bellingham, Wash.)
|July 6, 2023
PubMed
Summary

Serial dependence, a bias toward previously seen images, affects mammogram perception. This study used realistic, AI-generated mammograms to show this bias can lead to approximately 7% of categorization errors in clinical tasks.

Keywords:
generative adversarial networksradiological screeningserial dependencevisual search

More Related Videos

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands
10:59

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

Published on: July 26, 2014

14.5K
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

15.7K

Related Experiment Videos

Last Updated: Jul 24, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.9K
Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands
10:59

Reconstruction of 3-Dimensional Histology Volume and its Application to Study Mouse Mammary Glands

Published on: July 26, 2014

14.5K
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
14:27

Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data

Published on: June 26, 2013

15.7K

Area of Science:

  • Cognitive psychology
  • Medical imaging analysis
  • Artificial intelligence in healthcare

Background:

  • Human perception is prone to biases, particularly serial dependence, where recent stimuli influence current judgments.
  • Previous studies on serial dependence used unrealistic stimuli, limiting their applicability to real-world clinical scenarios.
  • The potential impact of serial dependence on clinicians' interpretation of medical images, such as mammograms, remains an area of active investigation.

Purpose of the Study:

  • To investigate the presence and characteristics of serial dependence in the perception of realistic mammograms.
  • To utilize advanced artificial intelligence, specifically generative adversarial networks (GANs), to create controlled and naturalistic mammogram stimuli.
  • To assess whether serial dependence influences diagnostic accuracy in a simulated mammogram perception task.

Main Methods:

  • A generative adversarial network (GAN) was trained on mammograms from the Digital Database for Screening Mammography (DDSM).
  • The trained GAN generated 2940 realistic, simulated mammograms across 20 morph continuums.
  • Participants performed a standard serial dependence experiment, viewing GAN-generated mammograms and reporting their perceptions.

Main Results:

  • Serial dependence was consistently observed across all naturalistic GAN-generated mammogram morph continuums.
  • Perceptual judgments of mammograms were significantly biased by previously viewed GAN-generated mammograms.
  • On average, approximately 7% of perceptual decisions exhibited categorization errors influenced by this serial dependence bias.

Conclusions:

  • Serial dependence demonstrably affects the perception of realistic, AI-generated mammograms.
  • The findings suggest that serial dependence is a potential contributing factor to decision errors in medical image interpretation.
  • This research highlights the importance of considering cognitive biases in the development of AI-assisted diagnostic tools and clinical workflows.