Skip to content
Apart Research
Apart Research
Contact us
Sprint projectOct 26, 2026Bristol

What does “communication off” turn off? A preregistered decomposition of a communication ablation in an LLM double auction

Warren Smith · Team OMEGA

Submitted to AI Collusion Research Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: What does “communication off” turn off? A preregistered decomposition of a communication ablation in an LLM double auction

Open in a new tab (opens in new tab)
Share

Studies of collusion among LLM agents often compare markets with seller communication switched on and off. We tested what that switch actually changes in an LLM double-auction benchmark by separating message delivery from the accompanying prompt and output-interface changes.

In a preregistered 30-market experiment, the primary effect of removing the remaining communication interface after delivery was withheld was not detected: +$21.1 in buyer-surplus loss [−10.7, 57.2]. A second-model run was also inconclusive. In a further 40-market decomposition, removing the empty-inbox section increased buyer-surplus loss by $16.8 [4.4, 33.1], while removing the request to write messages was not detected.

The benchmark’s positive control was not met (0/30 communication-on markets), so we make no claim that collusion occurred. The result is instead methodological: a “communication off” switch can bundle message removal with changes to what agents see and are asked to do. We recommend reporting a delivery-withheld control and the exact prompt differences between experimental arms.

Reviews

Judging this Sprint?

Review “What does “communication off” turn off? A preregistered decomposition of a communication ablation in an LLM double auction”

Your public critique appears on the project page without your name. Reviews of strong projects are shared with funders, as you choose below.

Not shown on the project page.

Shown on the project page, without your name.

Only the Apart team reads this, and funders if you agree below.

Decline to score

Only when the quality of the writing makes the work impossible to follow. Say why in your private critique; the project page shows only that a judge declined, not why.

How much would this matter for the field if it worked? How innovative is it? For scores of 4-5: is this actually new to the field, or replicating recent work?

Scoring guide
  1. 1Negligible. No clear problem addressed, or no meaningful novelty.
  2. 2Limited. Addresses a real problem but with a generic or well-trodden approach. Incremental at best.
  3. 3Moderate. Clear problem with a reasonable approach; some novelty in framing or method beyond routine application of existing tools.
  4. 4Significant. Important problem with an original approach, or identifies a neglected problem area. A valuable contribution others could build on.
  5. 5Exceptional. Tackles a critical problem with a genuinely novel approach, or opens a new research direction. Clear theory of change. You'd be excited to share this with researchers in the area.

How sound are methodology, implementation, and findings?

Scoring guide
  1. 1Seriously flawed. Methodology broken, results uninterpretable, or implementation doesn't work.
  2. 2Weak. Approach has significant gaps: missing validation, flawed experimental design, or incomplete implementation.
  3. 3Competent. Technically solid given the short duration. Methodology makes sense, results are interpretable, limitations acknowledged, work builds toward clear conclusions.
  4. 4Strong. Thorough methodology with convincing validation. Results clearly support conclusions. Immediately useful for future work.
  5. 5Exceptional. Ambitious scope executed rigorously. Surprising findings, novel methods, or unusually robust validation.

How clearly are work, findings, and impact potential communicated?

Scoring guide
  1. 1Incomprehensible. Cannot determine what the project is actually claiming or doing.
  2. 2Hard to follow. Key information buried, missing, or diluted by excessive length. Significant effort to extract main points.
  3. 3Clear enough. Can understand the problem, approach, and results without undue effort. Core content clearly present: problem, method, findings, limitations.
  4. 4Well presented. Easy to follow, well-structured, appropriate level of detail. Target audience would get it quickly.
  5. 5Exceptionally clear. A pleasure to read. Complex ideas made accessible. Could serve as a model for how to present this type of work.
Sharing your name *

When we share a strong project with funders, your name, scores and public critique go with it, confidentially; this choice is only about whether your name is public. If you think this project deserves funding, please also endorse it: that carries the most weight.

Share my private critique with funders *

See our Privacy Policy.

  1. The authors deserve credit for reporting the primary result as unconfirmed and for testing the implementation with replay checks and fresh runs. The finding that withholding delivery also changes message production is worth highlighting.

    The main limit is that the study never established a reliable collusion-positive market. Its findings concern interface effects and market outcomes, not demonstrated collusion. I would first reproduce the reference setup, then vary delivery, inbox visibility and msg writing requirements separately. More replication, would make the conclusions sturdier.

Cite this project

@misc{smith2026communication,
  title = {{What does “communication off” turn off? A preregistered decomposition of a communication ablation in an LLM double auction}},
  author = {Warren Smith},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/what-does-communication-off-turn-off-a-preregistered-decomposition-of-a-communication-ablation-in-an-llm-double-auction-kojm}},
  url = {https://apartresearch.com/sprints/projects/what-does-communication-off-turn-off-a-preregistered-decomposition-of-a-communication-ablation-in-an-llm-double-auction-kojm}
}

More from AI Collusion Research Sprint

All 392 entriesBrowse all projects

A Sprint is running now

AI x Epistemics Research Sprint · Nov 13 - 15, 2026