<![CDATA[The low level of active student engagement in Physical Education (PJOK) lessons, particularly in terms of communication, cooperation, and participation, is a major issue that highlights the need for a more interactive and meaningful learning approach. This study aims to describe student interactions in Physical Education (PJOK) lessons through the implementation of a deep learning approach at SMP Negeri 2 Cikajang. The study used a descriptive qualitative method with 30 seventh-grade students from class VII-G and 1 Physical Education teacher selected through purposive sampling. Data were collected through observation of interaction patterns and questionnaires, then analyzed through data reduction, data presentation, and drawing conclusions. Data validity was ensured through triangulation of sources and methods. The results of the study indicate that the application of deep learning significantly enhances two-way and multi-directional interactions, with the highest prevalence of two-way interactions occurring during closing activities (30.77%). Students demonstrated positive responses, high intrinsic motivation, and felt safe to express their opinions. However, student attention to movement demonstrations remained low (mean 1.70). The conclusion of this study is that the deep learning approach is effective in enhancing active, communicative, and meaningful learning interactions. Teachers are advised to improve the quality of movement demonstrations, and further research is needed to test its effectiveness quantitatively.]]>
