Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Prototyping Digital Social Spaces through Metaphor-Driven Design: Translating Spatial Concepts into an Interactive Social Simulation

Yoojin Hong, Martina Di Paola, Braahmi Padmakumar, Hwi Joon Lee, Mahnoor Shafiq, Joseph Seering

TL;DR

This work tackles the scarcity of prototyping space for alternative online social architectures by introducing a metaphor-driven system that converts spatial metaphors into eight social attributes, which are then mapped to a three-level feature taxonomy and instantiated in an interactive sandbox populated by LLM-driven agents. A two-phase study comparing open and closed social spaces demonstrates that metaphors elicit distinct social expectations, with perceived authenticity tied to content flow, identity visibility, and narrative coherence rather than features alone. The findings highlight metaphor-driven design as a powerful, low-cost approach to explore, evaluate, and expand the design space for future social platforms, while also outlining limitations in metaphor specificity, agent realism, and study generalizability. The paper suggests future work on richer feature integration, dynamic space adaptation during use, and enhanced visual representations to better translate metaphor into believable social ecosystems.

Abstract

Social media platforms are central to communication, yet their designs remain narrowly focused on engagement and scale. While researchers have proposed alternative visions for online spaces, these ideas are difficult to prototype within platform constraints. In this paper, we introduce a metaphor-driven system to help users imagine and explore new social media environments. The system translates users' metaphors into structured sets of platform features and generates interactive simulations populated with LLM-driven agents. To evaluate this approach, we conducted a study where participants created and interacted with simulated social media spaces. Our findings show that metaphors allow users to express distinct social expectations, and that perceived authenticity of the simulation depended on how well it captured dynamics like intimacy, participation, and temporal engagement. We conclude by discussing how metaphor-driven simulation can be a powerful design tool for prototyping alternative social architectures and expanding the design space for future social platforms.

Prototyping Digital Social Spaces through Metaphor-Driven Design: Translating Spatial Concepts into an Interactive Social Simulation

TL;DR

This work tackles the scarcity of prototyping space for alternative online social architectures by introducing a metaphor-driven system that converts spatial metaphors into eight social attributes, which are then mapped to a three-level feature taxonomy and instantiated in an interactive sandbox populated by LLM-driven agents. A two-phase study comparing open and closed social spaces demonstrates that metaphors elicit distinct social expectations, with perceived authenticity tied to content flow, identity visibility, and narrative coherence rather than features alone. The findings highlight metaphor-driven design as a powerful, low-cost approach to explore, evaluate, and expand the design space for future social platforms, while also outlining limitations in metaphor specificity, agent realism, and study generalizability. The paper suggests future work on richer feature integration, dynamic space adaptation during use, and enhanced visual representations to better translate metaphor into believable social ecosystems.

Abstract

Social media platforms are central to communication, yet their designs remain narrowly focused on engagement and scale. While researchers have proposed alternative visions for online spaces, these ideas are difficult to prototype within platform constraints. In this paper, we introduce a metaphor-driven system to help users imagine and explore new social media environments. The system translates users' metaphors into structured sets of platform features and generates interactive simulations populated with LLM-driven agents. To evaluate this approach, we conducted a study where participants created and interacted with simulated social media spaces. Our findings show that metaphors allow users to express distinct social expectations, and that perceived authenticity of the simulation depended on how well it captured dynamics like intimacy, participation, and temporal engagement. We conclude by discussing how metaphor-driven simulation can be a powerful design tool for prototyping alternative social architectures and expanding the design space for future social platforms.

Paper Structure

This paper contains 59 sections, 2 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Design Constraints leading Harms in Existing Social Media Platforms
  4. The Role of Spatial Metaphor in Rethinking Social Media
  5. Simulation as a Method for Social Media Prototyping
  6. System Design
  7. Converting a Metaphor into Social Attributes
  8. Translating metaphor into structured description
  9. Translating structured description into platform features
  10. Generation of an Interactive Social Platform
  11. Level 1: Layout Structure
  12. Level 2: Component Composition
  13. Level 3: Element Configuration and Behavioral Logic
  14. Populating LLM-Driven Social Agents Derived from User Metaphors
  15. Populating Social Agents and Network Construction
  16. ...and 44 more sections

Figures (2)

  • Figure 1: Overview of system pipeline, user study process, and research questions. This figure illustrates the end-to-end workflow of our study. The System Development phase (top) begins with user-provided spatial metaphors, which are expanded into structured metaphor descriptions (1S), translated into platform features via a 3-level taxonomy (2S), and instantiated as interactive social media simulations populated with LLM-driven agents by composing modular UI components into a simulation environment (3S). The User Study Process (bottom) evaluates each phase to address three research questions through: user-authored social attribute descriptions of the metaphor(1U), feature alignment ratings (2U), and post-interaction feedback (3U).
  • Figure 2: Results of participants' feature alignment ratings for open (green) vs closed (blue) bars. (A) shows the number of participants who identified each feature as aligned with their metaphor. (B) shows the number who identified each feature as misaligned.