Lottery and Sprint Arcade: Enabling Player-Driven Game Editing with Generative AI

Jul 16, 2026

Upload by Rowan

Large language models (LLMs) are changing game creation from offline generation to play-driven modification via natural-language interaction. The document presents a voice-command game editing system where spoken instructions are interpreted by an LLM and converted into structured updates of internal configuration parameters. A retro Space Invaders–style arcade game provides about 100 editable fields across mechanics, visuals, interactions, and audio, enabling iterative play–edit–feedback cycles. A user study analyzes experience, workload, and editing logs to examine interaction patterns and emergent editing strategies.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

Abstract

Introduction

Related Work

LLMs for Game Content Generation

arXiv :2607 . 10711v1 [ cs .HC] 12 Jul 2026 Lottery and Sprint Arcade: Enabling Player-Driven Game Editing with Generative AI Maya Grace Torii 1) Takahito Murakami 1) Yoichi Ochiai2) 1) Graduate School of Comprehensive Human Sciences, University of Tsukuba 2) R & D Centre for Digital Nature, University of Tsukuba {toriparu, takahito} (at) [digitalnature.slis.tsukuba.ac.jp](digitalnature.slis.tsukuba.ac.jp) Abstract Large language models (LLMs) are shifting game generation from offline automation toward play-driven modification through natural-language interaction. In this work, we present a play-driven game editing system that enables players to modify a retro Space Invaders – style arcade game through voice-based natural-language commands during play. Spoken instructions are interpreted by an LLM and translated into structured updates of internal configuration parameters, allowing iterative play − edit – feedback cycles in an invader-style game environment without exposing underlying system details. The game includes approximately 100 editable configuration fields controlling mechanics, visuals, interaction patterns, and audio behavior, enabling gameplay transformation through incremental parameter changes. To investigate how users experience play-driven AImediated editing (RQ1) and how emergent editing patterns relate to variations in player experience (RQ2), we conducted a user study combining subjective evaluations, workload measures, and log-based analysis of editing behavior. Participants were able to modify gameplay with generally positive experiences and moderate workload, and interaction outcomes did not strongly depend on prior programming experience. Editing-log analysis revealed distinct experiential tendencies: adjustments to immediately perceptible parameters were associated with higher usability, whereas edits affecting core gameplay structures were more closely associated with enjoyment. Post-session reflections further identified diverse editing strategies, including exploratory experimentation, goaldriven structural modification, and iterative parameter tuning. These findings demonstrate that voice-driven editing can support accessible, play-driven human – AI co-creation within a structured invader-style arcade game environment. Author-prepared manuscript. The version of record was published in The Journal of the Society for Art and Science, Vol. 25, No. 2, pp. 12:1–12:14, 2026 . 1 Introduction Digital game generation has been explored for many years through procedural content generation (PCG), algorithmic content generation (ACG), and machine learning – based approaches. These studies have demonstrated that games, levels, and rules can be produced by computational systems [1, 2] . More recently, large language models (LLMs) have introduced a new paradigm. Instead of generating complete games offline, LLM-based systems allow users to interact through natural language and iteratively modify content during use. LLMs shift game generation from offline automation to play-driven modification and co-creation through natural-language interaction [3, 4, 5] . In practice, AI-mediated editing may not simply reshape an existing game but lead to new forms of play or entirely different game experiences [6] . In this work, we present a play-driven game editing system that enables players to modify a retro Space Invaders – style arcade game through voice-based natural-language interaction during play. The system builds on our previous “Lottery and Sprint” method [7], a board game creation methodology that combines an LLM-based agent with the structured Design Sprint framework to support human – AI co-creation. The method was designed to make game design accessible to non-expert users by guiding iterative idea generation and refinement through AI-assisted workflows. Although Lottery and Sprint demonstrated that beginners could collaboratively design games with AI support, the process remained slow b

FAQs

What did the user study reveal about user experience and editing patterns?

Participants reported generally positive experiences with moderate workload, and outcomes were not strongly dependent on prior programming experience. Editing-log analysis linked usability with changes to immediately perceptible parameters, while enjoyment was more associated with edits affecting core gameplay structures.

Jul 16, 2026

Upload by Rowan