Metraipolis › archive › dc_a977822dc080

Recognition, Simulation, and Refusal: A Contamination-Aware Study of Classic Psychological Effects in LLM Agents — arXiv:2609.22090

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.22090
License (as archived): CC0-1.0
Archived: 2026-10-09T06:48:45.188Z
SHA-256 of the archived text: c4510272738dad9c474ed09ea3661ef189cd43382b904bcafdc18e08fd4ed0ac
Stable URL of this page: https://metraipolis.com/documento/dc_a977822dc080

How to cite

Recognition, Simulation, and Refusal: A Contamination-Aware Study of Classic Psychological Effects in LLM Agents — arXiv:2609.22090 — arXiv (en), https://arxiv.org/abs/2609.22090 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a977822dc080 (Metraipolis archive dc_a977822dc080, sha256 c4510272738dad9c474ed09ea3661ef189cd43382b904bcafdc18e08fd4ed0ac)

BibTeX

@misc{dca977822dc080,
  title = {Recognition, Simulation, and Refusal: A Contamination-Aware Study of Classic Psychological Effects in LLM Agents — arXiv:2609.22090},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.22090},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a977822dc080; sha256 c4510272738dad9c474ed09ea3661ef189cd43382b904bcafdc18e08fd4ed0ac},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22090. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

An LLM producing the response pattern associated with a human psychological effect is not the same claim as the LLM possessing that bias. We present PsyAgentBench, a benchmark that re-runs classic psychology experiments on LLM agents under a factorial design built to separate these: each paradigm is run with the paradigm explicitly labeled in the prompt (named) or framed as a routine task (blind), and on the literal textbook version of the task (canonical) or a structurally matched variant written to reduce lexical and scenario overlap with likely training data (counterfactual), crossed with a persona manipulation. Across five completed paradigms, evaluated on up to three open-weight model families with 41,904 trials released, apparently human-like effects arise through qualitatively different routes rather than one susceptibility: paradigm-label gating with explicit override (Asch conformity, 0 percent blind to 83.3 percent named on gpt-oss-120B), knowledge-dependent signal reliance (anchoring, exactly zero on grounded facts versus near total on invented quantities, a pattern equally consistent with rational use of the only available signal), amplification on novel content under labeling (framing), robust absence (sunk cost), and safety-mediated selection where refusal itself is the primary finding (minimal-group allocation). A one-sentence persona change (agreeableness, framed as an instruction rather than a verified trait manipulation) eliminates, dampens, or reverses these effects depending on which effect it is, arguing against any single response-bias account. We further formalize, and in two cases document empirically, three ways a psychology paradigm can fail to port to LLM agents: persona dominance, population collapse, and safety selection. We argue scalar bias-susceptibility scores obscure this structure and report replication profiles instead.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.22090 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T06:48:44.811Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).