A Literature Review: Which, How and What for the Use of Artificial Intelligence in Gamification

This literature review examines how artificial intelligence is applied within gamification and serious games in educational settings, drawing on 22 journal articles and conference papers published in English between 2019 and 2024. The 22 studies were selected from an initial pool of 1,163 records retrieved from Summon 2.0, Scopus, and IEEE Xplore using Boolean keyword searches, then filtered through defined inclusion and exclusion criteria. The review's stated aim is to identify trends in how AI supports gamified learning and to provide guidance for researchers, educators, developers, and policymakers.

The analysis organises AI applications into four dimensions: personalisation, adaptive difficulty, feedback and motivation, and predictive analytics with data-driven insights. Across these areas, the reviewed studies report that machine learning and related techniques can tailor challenges, rewards, and feedback to individual learners, dynamically adjust task difficulty to keep learners within an appropriate challenge zone, deliver real-time responses to learner actions, and forecast engagement, dropout risk, or task completion. The authors conclude that these applications are associated with improved learning outcomes and participation, with the strongest evidence concerning personalised and adaptive designs.

Several limitations shape how these findings should be read. The review relies on a qualitative, keyword-driven selection process without a stated quality appraisal framework or formal synthesis method, and many of the mappings between individual studies and the four dimensions are described in general terms. Some passages acknowledge that certain studies do not explicitly address the concepts assigned to them, indicating that the categorisation is interpretive rather than strictly empirical. The reference list also contains inconsistencies in citation years and author attributions, which affects the traceability of individual claims.