HXP-1: The Historical Experience Engine — A Multi-Agent Platform for Customer-Led Exploration of Reconstructed Past Worlds

Read full text
Authors: James Otto Danenberg
Cited by: 2
References: 4

Abstract

This paper defines the Historical Experience Engine: a platform for customer-led exploration of reconstructed past worlds that is neither a game (no win conditions, no scoring), nor a tour (full user agency), nor interactive fiction (physically and economically responsive world model), nor a chatbot (spatially embodied, multi-agent inhabited). The system is defined by a six-property cluster: no formal win state, strong user agency, a responsive world model, NPC social simulation, historically grounded reconstruction, and evaluation by presence rather than performance. The customer role is the flâneur — an observer-participant who moves through the world with embodied presence and reflective distance, governed by attention rather than objectives. A systematic survey of eleven commercial products and academic projects confirms that the five-pillar intersection (historically grounded, NPC-inhabited, simulation-maintained, customer-led, domain-agnostic) is currently vacant: existing systems satisfy one to three pillars but none satisfies all five. The platform architecture derives from the Adaptive Matrix Game backend, with the game layer removed and the simulation layer strengthened: procedural reconstruction, Regional RAG knowledge, certainty-gradient rendering, and multi-agent social simulation transfer; scoring, objectives, and deployment mechanics are replaced by narrative agency, domain parameterisation, and presence-centred evaluation. An evaluation framework is proposed spanning presence, emotional response, empathy, historical understanding, and cultural appropriateness. Future directions include therapeutic applications, heritage preservation, cultural tourism, and educational deployment.

Fields of Study & Subjects

Artificial Intelligence in Games Digital Games and Media 3D Surveying and Cultural Heritage

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
7 May 2026
Language
EN
Document type
Preprint