<![CDATA[Persistently low learning interest among Indonesian primary-school pupils in Islamic Religious Education (Pendidikan Agama Islam, PAI) constrains the depth of moral and cognitive engagement that the subject is intended to develop. Although Wordwall, a browser-based gamified-learning platform, has been examined in thematic, science, and mathematics instruction, its behavioral effect within the PAI context of Integrated Islamic Primary Schools (SDIT) remains underexplored. This study examined whether structured Wordwall use raises learning-interest indicators across two action-research cycles. Following the Kemmis and McTaggart spiral model, planning, action, observation, and reflection phases were enacted with 19 fourth-grade pupils (5 male, 14 female) at SDIT Mutiara Hati, Tebo Regency, during the even semester of 2024/2025. Wordwall activities (matching pairs, true-or-false, gameshow quiz) were embedded into PAI lessons on Praiseworthy Conduct (Akhlak Terpuji). Three behavioral indicators attention, active participation, and feelings of enjoyment were rated by two co-observers using a four-point rubric (±1 scale point inter-rater tolerance), supported by semi-structured interviews and photographic documentation. Quantitative scores were summarised as percentages; qualitative records were analysed through reduction, display, and verification. Mean learning interest rose from 38.0% (poor) at baseline to 56.5% (sufficient) after Cycle I and 81.0% (very good) after Cycle II, an absolute gain of 43.0 percentage points. All three indicators improved monotonically, with attention exhibiting the largest single-cycle gain (26.5 points, Cycle I). Wordwall-based gamified instruction is therefore an effective behavioural intervention for primary-school PAI in resource-constrained settings, and merits replication on larger samples and across different content domains]]>
