Designing a Large Language Model-Based AI System for Dynamic Difficulty Adjustment in Digital Games

Onur Aşkın

Makale detayları

Designing a Large Language Model-Based AI System for Dynamic Difficulty Adjustment in Digital Games

Yazarlar:

Onur Aşkın

Yayıncı:

Istanbul University

Ana Başlıklar:

Bilgi Sistemleri Kullanıcı Deneyimi Tasarımı ve Geliştirme | Grafikler | Artırılmış Gerçeklik ve Oyunlar (Diğer) | İnsan Bilgisayar Etkileşimi | Makine Öğrenme (Diğer) | Doğal Dil İşleme | Tasarım (Diğer)

Dergi:

Acta Infologica Vol.9 Issue 1, pp.183-207

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001354884

pISSN:

2602-3563

Anahtar kelimeler:

Özet:

 This study investigates how large language models LLMs can serve as dynamic agents in game-based interactions by comparing two prototypes of a color-guessing game. One model Cohere Command operates on a zero-shot prompt-based mechanism, while the other FLAN-T5 is fine-tuned on a semantically structured dataset. A total of 20 participants were divided into two experimental groups to evaluate the models ability to generate semantically coherent yesno questions, maintain flow, and perform accurate predictions. Quantitative data, including session durations, number of interactions, and AI outputs, were analyzed, along with a post-game user experience survey grounded in Flow Theory. Results show that while both systems achieved task completion, the fine-tuned FLAN-T5 model significantly outperformed the other models in terms of semantic clarity, user engagement, and perceived fluency. The findings highlight the potential of LLM-based DDA systems in creating meaningful, adaptive player experiences and underscore the importance of semantic alignment and interaction transparency in game-based AI design.

 This study investigates how large language models LLMs can serve as dynamic agents in game-based interactions by comparing two prototypes of a color-guessing game. One model Cohere Command operates on a zero-shot prompt-based mechanism, while the other FLAN-T5 is fine-tuned on a semantically structured dataset. A total of 20 participants were divided into two experimental groups to evaluate the models ability to generate semantically coherent yesno questions, maintain flow, and perform accurate predictions. Quantitative data, including session durations, number of interactions, and AI outputs, were analyzed, along with a post-game user experience survey grounded in Flow Theory. Results show that while both systems achieved task completion, the fine-tuned FLAN-T5 model significantly outperformed the other models in terms of semantic clarity, user engagement, and perceived fluency. The findings highlight the potential of LLM-based DDA systems in creating meaningful, adaptive player experiences and underscore the importance of semantic alignment and interaction transparency in game-based AI design.

 This study investigates how large language models LLMs can serve as dynamic agents in game-based interactions by comparing two prototypes of a color-guessing game. One model Cohere Command operates on a zero-shot prompt-based mechanism, while the other FLAN-T5 is fine-tuned on a semantically structured dataset. A total of 20 participants were divided into two experimental groups to evaluate the models ability to generate semantically coherent yesno questions, maintain flow, and perform accurate predictions. Quantitative data, including session durations, number of interactions, and AI outputs, were analyzed, along with a post-game user experience survey grounded in Flow Theory. Results show that while both systems achieved task completion, the fine-tuned FLAN-T5 model significantly outperformed the other models in terms of semantic clarity, user engagement, and perceived fluency. The findings highlight the potential of LLM-based DDA systems in creating meaningful, adaptive player experiences and underscore the importance of semantic alignment and interaction transparency in game-based AI design.

 This study investigates how large language models LLMs can serve as dynamic agents in game-based interactions by comparing two prototypes of a color-guessing game. One model Cohere Command operates on a zero-shot prompt-based mechanism, while the other FLAN-T5 is fine-tuned on a semantically structured dataset. A total of 20 participants were divided into two experimental groups to evaluate the models ability to generate semantically coherent yesno questions, maintain flow, and perform accurate predictions. Quantitative data, including session durations, number of interactions, and AI outputs, were analyzed, along with a post-game user experience survey grounded in Flow Theory. Results show that while both systems achieved task completion, the fine-tuned FLAN-T5 model significantly outperformed the other models in terms of semantic clarity, user engagement, and perceived fluency. The findings highlight the potential of LLM-based DDA systems in creating meaningful, adaptive player experiences and underscore the importance of semantic alignment and interaction transparency in game-based AI design.

Makale detayları

Designing a Large Language Model-Based AI System for Dynamic Difficulty Adjustment in Digital Games

Yazarlar:

Onur Aşkın

Yayıncı:

Istanbul University

Ana Başlıklar:

Bilgi Sistemleri Kullanıcı Deneyimi Tasarımı ve Geliştirme | Grafikler | Artırılmış Gerçeklik ve Oyunlar (Diğer) | İnsan Bilgisayar Etkileşimi | Makine Öğrenme (Diğer) | Doğal Dil İşleme | Tasarım (Diğer)

Dergi:

Acta Infologica Vol.9 Issue 1, pp.183-207

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001354884

pISSN:

2602-3563

Artificial Intelligence

Machine Learning

Digital Game Design

Dynamic Difficulty Adjustment

Sayı

Makale detayları

Designing a Large Language Model-Based AI System for Dynamic Difficulty Adjustment in Digital Games

Yazarlar: Onur Aşkın

Yayıncı: Istanbul University

Ana Başlıklar: Bilgi Sistemleri Kullanıcı Deneyimi Tasarımı ve Geliştirme | Grafikler | Artırılmış Gerçeklik ve Oyunlar (Diğer) | İnsan Bilgisayar Etkileşimi | Makine Öğrenme (Diğer) | Doğal Dil İşleme | Tasarım (Diğer)

Dergi: Acta Infologica Vol.9 Issue 1, pp.183-207

Belge Tipi: Makale

Yayın Tarihi: 6/2025

Belge Dili: İngilizce

Document ID (Org No.): 2001354884

pISSN: 2602-3563

Artificial Intelligence

Machine Learning

Digital Game Design

Dynamic Difficulty Adjustment

Sayı

Yazarlar:

Onur Aşkın

Yayıncı:

Istanbul University

Ana Başlıklar:

Bilgi Sistemleri Kullanıcı Deneyimi Tasarımı ve Geliştirme | Grafikler | Artırılmış Gerçeklik ve Oyunlar (Diğer) | İnsan Bilgisayar Etkileşimi | Makine Öğrenme (Diğer) | Doğal Dil İşleme | Tasarım (Diğer)

Dergi:

Acta Infologica Vol.9 Issue 1, pp.183-207

Belge Tipi:

Makale

Yayın Tarihi:

6/2025

Belge Dili:

İngilizce

Document ID (Org No.):

2001354884

pISSN:

2602-3563