<![CDATA[Problem-solving ability is one of the fundamental cognitive competencies that must be developed from an early age, because it is the basis for children's success in facing challenges in subsequent stages of education. This study aims to improve the problem-solving abilities of early childhood through the use of puzzle game technology. The study was conducted from April to May 2026 using the Classroom Action Research (CAR) method of the Kemmis and McTaggart model consisting of two cycles. Each cycle includes the stages of planning, implementation, observation, and reflection. The research subjects were 14 children in group B aged 5–6 years. The data collection instrument used an observation sheet that refers to the four indicators of problem-solving according to Polya, namely understanding the problem, making a plan, implementing the plan, and re-examining. Data were analyzed descriptively quantitatively and supported by paired sample t-tests. The results of the study showed a significant increase in each cycle. In the pre-action, the average score of children's problem-solving abilities was 52% (Starting to Develop). After being given intervention in cycle I, the average score increased to 74% (Developing as Expected), and in cycle II it increased to 89% (Developing Very Well). The highest improvements occurred in the indicators of implementing plans (from 55% to 91%) and understanding problems (from 45% to 85%). The results of the paired sample t-test showed a calculated t value of 6.847 > t table = 2.080, indicating a significant difference between children's problem-solving abilities before and after treatment. The conclusion of this study is that the use of puzzle game technology is effective in improving the problem-solving abilities of early childhood children at Makarti Kindergarten, PKO Seluma.]]>
