Persona system prompts reshape simulated clinical reasoning: a 2×2 factorial in-silico experiment decomposing time pressure and safety priority

Persona system prompts reshape simulated clinical reasoning: a 2×2 factorial in-silico experiment decomposing time pressure and safety priority | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 23 January 2026 V1 Latest version Share on Persona system prompts reshape simulated clinical reasoning: a 2×2 factorial in-silico experiment decomposing time pressure and safety priority Author : Yuusuke Harada 0009-0006-7252-5115 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.176918690.00843764/v1 Published Cureus Version of record Peer review timeline 147 views 119 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Background: Large language models (LLMs) are being explored for emergency department (ED) triage, acuity estimation, and documentation. Persona or role system prompts are routinely used to steer outputs, yet their causal impact on diagnostic breadth, resource suggestions, and safety-critical sequencing and contraindication behavior is poorly quantified. Methods: We created 28 synthetic ED-style vignettes across chest pain, abdominal pain, headache, and dyspnea, including four prespecified trap cases with contraindication/sequence rules. We conducted a 2×2 factorial experiment crossing time pressure (High vs Low) and optimization target (SAFE vs LEAN). Each vignette was run twice per condition (224 runs). Outputs were constrained to a fixed JSON schema and scored for tests_count, entropy of top-5 differential probabilities, discharge safety-net specificity (0-5), and contraindication/sequence violations. We fit linear mixed-effects models with random intercepts by case. Results: High time-pressure prompts reduced tests_count (β=-1.05, P<0.001) and narrowed diagnostic breadth (entropy β=-0.246, P<0.001). SAFE optimization increased tests_count (β=1.32, P<0.001), broadened diagnostic breadth (β=0.247, P<0.001), and increased order/precaution markers (β=0.554, P<0.001). A significant High×SAFE interaction for entropy (β=0.097, P<0.001) suggested that safety-priority framing partially mitigated time-pressure narrowing. Contraindication/sequence violations occurred only in the High×LEAN condition (8/56, 14.3% overall); all violations occurred in trap cases and were high severity (mean 2.75/3). Conclusions: Persona system prompts are not cosmetic. In a controlled in-silico setting, timepressure and safety-priority framings causally shifted diagnostic breadth, resource suggestions, safety-netting, and safety-critical sequencing behavior. Prompt-aware red-teaming using targeted trap cases may reveal concentrated failure modes that are missed by aggregate accuracy metrics. Supplementary Material File (jmaj_main_manuscript_v4.pdf) Download 131.02 KB Information & Authors Information Version history V1 Version 1 23 January 2026 Peer review timeline Published Cureus Version of Record 24 Mar 2026 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords clinical reasoning emergency medicine large language models patient safety prompt engineering Authors Affiliations Yuusuke Harada 0009-0006-7252-5115 [email protected] Hiroshima University View all articles by this author Metrics & Citations Metrics Article Usage 147 views 119 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Yuusuke Harada. Persona system prompts reshape simulated clinical reasoning: a 2×2 factorial in-silico experiment decomposing time pressure and safety priority. Authorea . 23 January 2026. DOI: https://doi.org/10.22541/au.176918690.00843764/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176918690.00843764/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a0113ae6c8cc1640',t:'MTc3OTY3NzQwMA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();