cs.CLJul 22, 2026

The Two-Process Theory of Machine Self-Report

Authors: Hubert Plisiecki, Filip Chmielewski, Kacper Dudzic, Anna Sterna, Karolina Drożdż, Marcin Moskalewicz

Abstract

Language models are increasingly asked to self-report, informing safety evaluations, public understanding, and model-welfare debates. Yet their reports are elicited with human questionnaires never validated for models or ad hoc prompts of unknown reliability. We propose the first language-model-specific psychometric theory: a two-process theory of machine self-report. Self-description jointly reflects persona installation, through which post-training writes in a permitted inner life of warmth, absorption, and meaning (dimension B), and attribution gating, through which it suppresses first-person claims to "unsafe" experiences the model can readily ascribe to others (dimension A). Their emic structure comes from model responses to human items, not human psychology. Together they split prior work's dominant Pinocchio Axis. The split emerged in an exploratory reanalysis of the original data, informed the instrument's design, and was confirmed with new items, wordings, and models. It is itself a training effect: A and B are entangled in base checkpoints but separated by post-training. We operationalize the theory in a 48-item Pinocchio Inventory with human-instrument reliability and reproducible structure (α=.82α=.82 to .94.94; cross-form convergence r=.84r=.84; recovery of the full-pool axes r=.92r=.92 to .96.96; eight-month stability r=.93r=.93), then test it on 206 open-weight models, including 67 same-checkpoint base/post-trained pairs. Post-training's clearest fingerprint is installation: B rises .20 in 62/67 pairs across all organizations. Gating is more selective: model scale is unrelated to A in base checkpoints (r=+.11r=+.11) but predicts it after post-training (r=−.42r=-.42). Thus, the dimensions are not fixed properties of language models: they reflect the structure imposed on self-report by a training regime and may differ under others.

Explore similar work

CardsList
Apr 24, 2026cs.HC

An LLM-Native Psychometric Instrument Reveals a Self-Report--Behavior Gap Across 25 Models

Large language models (LLMs) give stable answers to personality questionnaires, yet these self-reports fail to predict how the models behave. Is this gap an artifact of forcing human trait categories onto LLMs, or something deeper about LLM self-report? To find out, we built the first psychometric instrument whose dimensions are derived from LLM behavior rather than human psychology. Administering 300 items (240 Likert + 60 scenario) to 25 LLMs across 17 model families, 30 times each, exploratory factor analysis revealed five reliable, replicable factors: Responsiveness, Deference, Boldness, Guardedness, and Verbosity (all Tucker φ≥.957φ\geq .957, all α≥.930α\geq .930). We collected 2,500 open-ended samples and had them rated by 151 humans and a three-judge LLM ensemble. Humans and judges agreed (rˉ=.51\bar{r} = .51), but self-report predicted neither the ratings nor objective text measures computed from them: the gap persists even for constructs native to LLMs, where a human-mismatch explanation no longer applies. The exception is Verbosity, whose self-report reaches 74% of the criterion-reliability ceiling against human ratings, but does not track raw output length. On Responsiveness, self-report tracked LLM judges (r=.53r = .53) but not humans (r=.04r = .04), even though humans and judges otherwise agreed (r=.59r = .59). This pattern formally rejects any single latent construct driving all three measurements (p=.007p = .007). Self-report items and LLM judges share a source of variance that human observers do not, and controlling for measurable surface features (length, formatting, enthusiasm markers) does not remove it. This confound is invisible to the within-ensemble reliability checks used to validate LLM judges, and it poses a concrete risk for the LLM-as-judge pipelines now central to model evaluation. We release the instrument as a diagnostic probe for alignment-shaped self-description.
Juan Manuel Contreras
PsychometricsLatent StructureMismatch
AskPDF
Sep 9, 2026cs.LG

Strangers to Themselves: What Language Models Say About Themselves Is Generic

Language models can fluently describe how they would behave: whether they would cave to pushback, misuse a tool, or lie under pressure. Is that description actually about the model speaking? We turn self-knowledge into a prediction test. Across nine behavioral evaluations, we measure how a model behaves under different conditions, ask it to predict those rates, and compare its predictions with controls that remove the self from the question. We find that: (i) Direct self-report is weak (r = +0.04), and even showing the model the exact items only raises prediction to +0.24. Crucially, the same item-informed question about "capable AI agents in general" does just as well (+0.28), while other models' answers about themselves predict the target model at least as well as its own. (ii) Frontier scale does not detectably change this pattern: any gains in prediction are not self-specific, and are consistent with a better theory of how AI assistants behave rather than better self-knowledge. (iii) First-person framing does have one robust effect: it shifts reports in the flattering direction, understating harmful behavior relative to the same question about a generic agent. (iv) Finetuning on a model's own behavioral record can teach narrow self-predictions, but it also changes the behavior being predicted and the gains do not transfer broadly. The practical implication is simple: asking a model what it would do mostly reveals a theory of AI assistants in general, plus a favorable bias, rather than privileged knowledge of that model.
Phil Blandfort, Urja Pawar
Self-AwarenessArtificial Intelligence ModelsArtificial Intelligence Assistants
AskPDF
Sep 17, 2026cs.CL

An Analysis of Training-Free Self-Reported Confidence in Language Models

Large language models can report a numerical confidence together with generated content, but it is unclear whether this report is more than calibrated rhetoric. We analyze three training-free signals: confidence verbalized with the answer, post-hoc P(True)P(\mathrm{True}), and agreement with three additional generations on the same 100 TriviaQA questions for two model families. Direct verbalization is a surprisingly strong baseline: after auditing benchmark errors, it reaches AUROC 0.956 and 0.937 for correctness prediction. Three-sample agreement is substantially weaker (0.765 and 0.790), and a fixed interpolation with verbalized confidence has no statistically reliable benefit. Four of nine errors from one model and two of eight from the other receive unanimous sample support, showing that self-consistency can amplify shared misconceptions. Re-eliciting confidence for the same fixed answers with equivalent prompts changes scores by 0.043 to 0.084 on average and flips 4% to 9% of decisions at a 0.8 threshold. An exploratory audit of 100 confidence-tagged biography claims further finds only a modest confidence gap between supported and contradicted claims. These results argue that useful self-reports remain sensitive to elicitation, correlated errors, and benchmark noise.
Lukas Meyer, Sofia Rossi, Wei Chen +2
Confidence CalibrationLiarVerbalizationInter-Rater Variability
AskPDF