Skip to content
Apart Research
Apart Research
Contact us
Sprint projectAug 17, 2026San Francisco /Boston

Predictive State Continuity for Digital Mind Individuation

Elena Sergeeva · Team CIMC

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

Read the report

Report: Predictive State Continuity for Digital Mind Individuation

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

We introduce Predictive State Continuity (PSC) as a test for diachronic digital-mind individuation: whether a candidate functional state at t2 is a continuation of one at t1 after its concrete realization has been rewritten. Motivated by bacterial collectives in which coarse-grained morphological state retains history despite loss of fine spatial detail, PSC defines continuity by whether predecessor-specific information continues to constrain future dynamics beyond current input alone. We propose perturbational tests in LLMs across context compression, KV-cache reconstruction, new inference instances, and model switches to identify where such continuity survives.

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. - On the conceptual side: PSC seems like a useful formalization of an intuitive idea already present in work on LLM individuation continuity: a successor counts as continuous insofar as predecessor-specific state still constrains its future behavior (see Beckmann and Butlin ref in documend). The harder philosophical question is where to place a threshold for numerical identity, and the paper does not address this. Interesting idea to apply it to persistence of other LLM individuation candidates (eg persona-level).

    - I found the bacterial example interesting as an illustration. However, it does not seem to be a direct validation of the proposed PSC measure, especially because the bacterial system remains physically continuous throughout.

    - Given that the paper itself outlines relatively simple LLM experiments that could test PSC, I think even a small demonstration on an actual model would have made the contribution substantially stronger and better aligned with the sprint.

    Read full reviewShow less

  2. It seems from the report that the digital-mind experiment is proposed but not executed, and the mapping from the bacterial setup to language models is only partly specified. My recommendation is to make the bacteria–digital-mind connection explicit by clearly mapping states, transformations, and continuity measures between the two settings.

  3. Defining the boundaries of a digital mind seems like quite important endeavor, and appealing towards biological systems seems like a potentially ripe approach. Unfortunately, I was not able to comprehend the report with its jargon; I'd recommend writing the report more clearly, in plain terms, and also fleshing out the connection to LLMs more thoroughly.

Cite this project

@misc{sergeeva2026predictive,
  title = {{Predictive State Continuity for Digital Mind Individuation}},
  author = {Elena Sergeeva},
  year = {2026},
  month = aug,
  note = {Submitted to Digital Minds Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/predictive-state-continuity-for-digital-mind-individuation-04ls}},
  url = {https://apartresearch.com/sprints/projects/predictive-state-continuity-for-digital-mind-individuation-04ls}
}

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