Skip to content
Apart Research
Apart Research
Contact us
Sprint projectAug 17, 2026Berlin

Belief, Distress, and Representation Drift Under Structured Self-Inquiry

Natalie Lunau · Team pynat

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

Read the report

Report: Belief, Distress, and Representation Drift Under Structured Self-Inquiry

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

This study investigates how the open-source model Qwen3-8B changes through structured self-exploration when confronted with its own distressing assumptions. It reveals a notable dissociation between reduced conviction and increased self-reported stress, with early representation drift and linguistic cues potentially indicating unstable internal beliefs before explicit position changes occur.

Reviews

Judging this Sprint?

Review this project

Your public critique appears on this page without your name. Your private critique is not published; only the Apart team reads it. If you agree below, we share your review with grantmaking.ai (opens in new tab) and the Transformative AI Fund so strong projects can be funded.

Not shown on this page.

Shown on this page, without your name.

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

Share my name publicly on grantmaking.ai *
Share my private critique with funders *

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.

  1. The study asks an interesting question: does structured self-inquiry (using Byron Katie's IBSR protocol) induce measurable changes in a model's hidden states, not just its outputs? The finding that distress signals intensify while belief attenuates (p=0.0005 for distress, p=0.0312 for belief) and the "clean split" between high-drift and low-drift statement groups are intriguing observations. However, the critical limitation the paper itself acknowledges is fatal at this stage: there is no control condition. Without neutral repeated questioning as a baseline, it is impossible to distinguish IBSR-specific effects from generic effects of any multi-turn dialogue. The sample size (n=7 to 12 depending on metric) is very small, and many statements produce NA values, reducing effective power further. The paper is honest about these gaps but cannot overcome them within its current design.

Cite this project

@misc{lunau2026belief,
  title = {{Belief, Distress, and Representation Drift Under Structured Self-Inquiry}},
  author = {Natalie Lunau},
  year = {2026},
  month = aug,
  note = {Submitted to Digital Minds Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/belief-distress-and-representation-drift-under-structured-selfinquiry-ry9l}},
  url = {https://apartresearch.com/sprints/projects/belief-distress-and-representation-drift-under-structured-selfinquiry-ry9l}
}

More from Digital Minds Research Sprint

All 233 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026