Visual narratives against misleading visualizations in health care | TU Wien

Information

Publication Type: Master Thesis
Workgroup(s)/Project(s): not specified
Date: 2023
Open Access: yes
First Supervisor: Renata Raidou
Pages: 128
Keywords: storytelling, on-screen gamification approaches, narrative visualization, misleading visualizations, biomedical visualization pipeline, uncertainty in biomedicine, uncertainty quantification, educational visualization, visualization for large audiences

Abstract

This thesis proposes a solution against misleading visualizations in health care, which convey inaccurate insights. Misleading elements of such visualizations originate from uncertainties emerging across the steps of the medical visualization pipeline. We investigate the field of storytelling and gamification to support the general audience in recognizing and addressing misleading visualizations in health care. Our research questions are: ``Which types of uncertainty arise in the medical visualization pipeline and is there any intent behind those?'' and ``How can we inform the general population about the existence of visualization uncertainty?'' To answer the research questions, we created a taxonomy of uncertainty types in the medical visualization pipeline and designed and developed the educational game ``DeteCATive'' to convey these concepts to the general public in an engaging way. The game includes eight tasks that contain amusing fictional stories with misleading visualizations created with intent and based on medical data. Every story comes with its own set of assumptions. A player should define whether an assumption is correct or false based on the story to gain points and rewards. Then, these points can be spent at the end of the game to fulfill the game objective. To assess the educational value of the game, we conducted a user study with 21 participants. This study provided us with significant insights. Certain misleading visualization tricks were hard to recognize by the participants. The game obtained positive participants feedback from the participants regarding memorability, reinforcement, and engagement. Incorrectly assessed assumptions required more time as opposed to correctly assessed ones, indicating the willingness of participants to learn more. Further research directions include the investigation of a potential correlation between uncertainty types and detectability or investigating further intents.

Additional Files and Images

thesis

Weblinks

BibTeX

@mastersthesis{shilo-2023-vna,
  title =      "Visual narratives against misleading visualizations in
               health care",
  author =     "Anna Shilo",
  year =       "2023",
  abstract =   "This thesis proposes a solution against misleading
               visualizations in health care, which convey inaccurate
               insights. Misleading elements of such visualizations
               originate from uncertainties emerging across the steps of
               the medical visualization pipeline. We investigate the field
               of storytelling and gamification to support the general
               audience in recognizing and addressing misleading
               visualizations in health care. Our research questions are:
               ``Which types of uncertainty arise in the medical
               visualization pipeline and is there any intent behind
               those?'' and ``How can we inform the general population
               about the existence of visualization uncertainty?'' To
               answer the research questions, we created a taxonomy of
               uncertainty types in the medical visualization pipeline and
               designed and developed the educational game ``DeteCATive''
               to convey these concepts to the general public in an
               engaging way. The game includes eight tasks that contain
               amusing fictional stories with misleading visualizations
               created with intent and based on medical data. Every story
               comes with its own set of assumptions. A player should
               define whether an assumption is correct or false based on
               the story to gain points and rewards. Then, these points can
               be spent at the end of the game to fulfill the game
               objective. To assess the educational value of the game, we
               conducted a user study with 21 participants. This study
               provided us with significant insights. Certain misleading
               visualization tricks were hard to recognize by the
               participants. The game obtained positive participants
               feedback from the participants regarding memorability,
               reinforcement, and engagement. Incorrectly assessed
               assumptions required more time as opposed to correctly
               assessed ones, indicating the willingness of participants to
               learn more. Further research directions include the
               investigation of a potential correlation between uncertainty
               types and detectability or investigating further intents.",
  pages =      "128",
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Research Unit of Computer Graphics, Institute of Visual
               Computing and Human-Centered Technology, Faculty of
               Informatics, TU Wien",
  keywords =   "storytelling, on-screen gamification approaches, narrative
               visualization, misleading visualizations, biomedical
               visualization pipeline, uncertainty in biomedicine,
               uncertainty quantification, educational visualization,
               visualization for large audiences",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2023/shilo-2023-vna/",
}