Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

The Centers and Margins of Modeling Humans in Well-being Technologies: A Decentering Approach

Jichen Zhu, Pedro Sanches, Vasiliki Tsaknaki, Willem van der Maden, Irene Kaklopoulou

TL;DR

The paper critiques how current ML modeling of humans in well-being technologies encodes narrow, often cisgender, normed notions of the body and health. Using agential realism within a critical technical practice, it analyzes three case studies (Clue, a personalized social fitness app, and the Oura Ring) to reveal centralized assumptions and where they margin users’ diverse experiences. It then proposes a posthuman-centered ML framework that treats humans and environments as entangled agents, advocates alternative modeling approaches (e.g., plural temporalities, multi-agent dynamics, and diffraction-based design), and outlines design implications for more inclusive, adaptive wellbeing technologies. The work highlights the need for ongoing reflexivity, methodological pluralism, and cross-disciplinary collaboration to advance ML design that accommodates irregularities, transitions, and the co-constitutive nature of humans with technology and environment.

Abstract

This paper critically examines the machine learning (ML) modeling of humans in three case studies of well-being technologies. Through a critical technical approach, it examines how these apps were experienced in daily life (technology in use) to surface breakdowns and to identify the assumptions about the "human" body entrenched in the ML models (technology design). To address these issues, this paper applies agential realism to decenter foundational assumptions, such as body regularity and health/illness binaries, and speculates more inclusive design and ML modeling paths that acknowledge irregularity, human-system entanglements, and uncertain transitions. This work is among the first to explore the implications of decentering theories in computational modeling of human bodies and well-being, offering insights for more inclusive technologies and speculations toward posthuman-centered ML modeling.

The Centers and Margins of Modeling Humans in Well-being Technologies: A Decentering Approach

TL;DR

The paper critiques how current ML modeling of humans in well-being technologies encodes narrow, often cisgender, normed notions of the body and health. Using agential realism within a critical technical practice, it analyzes three case studies (Clue, a personalized social fitness app, and the Oura Ring) to reveal centralized assumptions and where they margin users’ diverse experiences. It then proposes a posthuman-centered ML framework that treats humans and environments as entangled agents, advocates alternative modeling approaches (e.g., plural temporalities, multi-agent dynamics, and diffraction-based design), and outlines design implications for more inclusive, adaptive wellbeing technologies. The work highlights the need for ongoing reflexivity, methodological pluralism, and cross-disciplinary collaboration to advance ML design that accommodates irregularities, transitions, and the co-constitutive nature of humans with technology and environment.

Abstract

This paper critically examines the machine learning (ML) modeling of humans in three case studies of well-being technologies. Through a critical technical approach, it examines how these apps were experienced in daily life (technology in use) to surface breakdowns and to identify the assumptions about the "human" body entrenched in the ML models (technology design). To address these issues, this paper applies agential realism to decenter foundational assumptions, such as body regularity and health/illness binaries, and speculates more inclusive design and ML modeling paths that acknowledge irregularity, human-system entanglements, and uncertain transitions. This work is among the first to explore the implications of decentering theories in computational modeling of human bodies and well-being, offering insights for more inclusive technologies and speculations toward posthuman-centered ML modeling.

Paper Structure

This paper contains 30 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. "Humans" in Human-Centered ML
  4. Posthuman Concepts in HCI and Design: A Focus on Decentering
  5. Decentering through Agential Realism
  6. Critical Technical Practice
  7. Methodology
  8. Selection of Case Studies
  9. Analysis and Methods in Each Case Study
  10. Three Case Studies on Designing ML Well-being Technologies
  11. Case Study 1 - Modeling of Bodies in Clue
  12. Context/Background
  13. Analysis to Reveal Embedded Assumptions
  14. Decentering Assumptions Through Agential Realism
  15. Case Study 2 - Modeling of Preferences in a Personalized Social Fitness App
  16. ...and 15 more sections

Figures (3)

  • Figure 1: A notification in Clue stating that comparisons to "normal" are based on global statistics (Case Study 1).
  • Figure 2: Screenshot of a personalized social comparison page with other users to motivate physical activities (Case Study 2).
  • Figure 3: Well-being transitions between readiness tiers and the readiness score assessments in the first autoethnographic entry (Case Study 3).