Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Can Generative Agent-Based Modeling Replicate the Friendship Paradox in Social Media Simulations?

Gian Marco Orlando, Valerio La Gatta, Diego Russo, Vincenzo Moscato

TL;DR

The paper investigates whether Generative Agent-Based Modeling (GABM) can reproduce the Friendship Paradox (FP) and its generalizations (GFP) in social media networks. It introduces a GABM framework with LLM-powered generative agents that possess profiles, memory, and reasoning, interacting via a retrieval-augmented workflow on two Twitter datasets. The experiments show FP and GFP emerge naturally, revealing a hierarchical pattern where infrequent connections to higher-activity or higher-influence agents drive the paradox, consistent with real networks. This work demonstrates GABM as a robust tool for analyzing global online social phenomena and offers a framework for studying misinformation and polarization dynamics in digital environments.

Abstract

Generative Agent-Based Modeling (GABM) is an emerging simulation paradigm that combines the reasoning abilities of Large Language Models with traditional Agent-Based Modeling to replicate complex social behaviors, including interactions on social media. While prior work has focused on localized phenomena such as opinion formation and information spread, its potential to capture global network dynamics remains underexplored. This paper addresses this gap by analyzing GABM-based social media simulations through the lens of the Friendship Paradox (FP), a counterintuitive phenomenon where individuals, on average, have fewer friends than their friends. We propose a GABM framework for social media simulations, featuring generative agents that emulate real users with distinct personalities and interests. Using Twitter datasets on the US 2020 Election and the QAnon conspiracy, we show that the FP emerges naturally in GABM simulations. Consistent with real-world observations, the simulations unveil a hierarchical structure, where agents preferentially connect with others displaying higher activity or influence. Additionally, we find that infrequent connections primarily drive the FP, reflecting patterns in real networks. These findings validate GABM as a robust tool for modeling global social media phenomena and highlight its potential for advancing social science by enabling nuanced analysis of user behavior.

Can Generative Agent-Based Modeling Replicate the Friendship Paradox in Social Media Simulations?

TL;DR

The paper investigates whether Generative Agent-Based Modeling (GABM) can reproduce the Friendship Paradox (FP) and its generalizations (GFP) in social media networks. It introduces a GABM framework with LLM-powered generative agents that possess profiles, memory, and reasoning, interacting via a retrieval-augmented workflow on two Twitter datasets. The experiments show FP and GFP emerge naturally, revealing a hierarchical pattern where infrequent connections to higher-activity or higher-influence agents drive the paradox, consistent with real networks. This work demonstrates GABM as a robust tool for analyzing global online social phenomena and offers a framework for studying misinformation and polarization dynamics in digital environments.

Abstract

Generative Agent-Based Modeling (GABM) is an emerging simulation paradigm that combines the reasoning abilities of Large Language Models with traditional Agent-Based Modeling to replicate complex social behaviors, including interactions on social media. While prior work has focused on localized phenomena such as opinion formation and information spread, its potential to capture global network dynamics remains underexplored. This paper addresses this gap by analyzing GABM-based social media simulations through the lens of the Friendship Paradox (FP), a counterintuitive phenomenon where individuals, on average, have fewer friends than their friends. We propose a GABM framework for social media simulations, featuring generative agents that emulate real users with distinct personalities and interests. Using Twitter datasets on the US 2020 Election and the QAnon conspiracy, we show that the FP emerges naturally in GABM simulations. Consistent with real-world observations, the simulations unveil a hierarchical structure, where agents preferentially connect with others displaying higher activity or influence. Additionally, we find that infrequent connections primarily drive the FP, reflecting patterns in real networks. These findings validate GABM as a robust tool for modeling global social media phenomena and highlight its potential for advancing social science by enabling nuanced analysis of user behavior.

Paper Structure

This paper contains 11 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methodology
  3. Generative Agents
  4. Simulation Workflow
  5. Experiments
  6. Datasets
  7. Experimental setup
  8. Measuring the Friendship Paradox
  9. RQ1: Do the FP and GFP manifest in GABM-based social media simulations?
  10. RQ2: Which social connections most contribute to experiencing the FP and GFP?
  11. Conclusion & Future Work

Figures (3)

  • Figure 1: Example of personality inference. LLMs analyze user's generated content to extract personality traits.
  • Figure 2: The simulation workflow begins with the Initialization Phase, setting up agents with distinct personalities. In the Operational Phase, agents autonomously decide their actions, such as posting or interacting with content. The Interaction Phase updates the simulation environment. The process repeats iteratively until the stop condition is met.
  • Figure 3: Proportion of agents experiencing mean follower superiority and mean followee superiority for each attribute across different restriction conditions.