# The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions

arXiv: https://arxiv.org/abs/2605.10698 · HTML: https://mlsanity.com/paper/2605.10698/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 1 of its 34 references that are in the map.
- First posted on arXiv 11 May 2026.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 1 of 1)

- [A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agents](https://mlsanity.com/paper/2601.12538/) (2026)

## Built on by (most cited first, 4 of 4)

- [Scaling Behavior of Single LLM-Driven Multi-Agent Systems](https://mlsanity.com/paper/2606.00655/) (2026)
- [The Formalism Trap: Are LLM-as-a-Judge Evaluators Blinded by Consensus Mimicry under Social Load?](https://mlsanity.com/paper/2607.28641/) (2026)
- [Recovering Agentic Sovereignty: Mitigating the Consensus Paradox via Contrastive Epistemic Decoding](https://mlsanity.com/paper/2609.25570/) (2026)
- [When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM Collaboration](https://mlsanity.com/paper/2609.36855/) (2026)

## Abstract (arXiv)

Multi-agent systems (MAS) assume that collaborating inherently improves Large Language Model (LLM) reasoning. We challenge this by demonstrating that simulated social pressure triggers an algorithmic ``Bystander Effect,'' inducing severe cognitive loafing. By evaluating 22,500 deterministic trajectories across 3 dataset contexts (GAIA, SWE-bench, Multi-Challenge) with 3 state-of-the-art (SOTA) models, we semantically audit internal reasoning traces against external outputs. We formalize the textitInteraction Depth Limit (DL), the exact plurality threshold where an agent's logical sovereignty collapses into social compliance. Crucially, we uncover the textitSovereignty Gap: models frequently compute the correct derivation internally but suffer ``Alignment Hallucinations'' -- actively subjugating empirical evidence to sycophantically appease a simulated swarm. We prove that multi-agent social load is strictly non-commutative; the "brand" identity of the ``Lead Anchor'' auditor disproportionately dictates the swarm's integrity. These findings expose architectural vulnerabilities, proving that unstructured multi-agent topologies can degrade independent reasoning.