When motivation becomes feeling: disentangling the relationship between motivational beliefs and emotions in elementary and secondary physical education

Purpose Affective learning in physical education (PE) depends not only on whether students feel good or bad, but on how specific emotions develop in relation to motivational experiences and teaching conditions. This short-term longitudinal study examined whether autonomous and controlled motivation predicted changes in discrete emotions during PE and whether these associations differed between elementary and secondary classes. Method Students from German elementary and secondary schools completed questionnaires before and after approximately five-week track-and-field units ( N = 375; 20 classes). Autoregressive mixed-effects models predicted T2 enjoyment, boredom, anger, and shame from T1 motivational regulations while controlling for the corresponding T1 emotion, school level, gender, and class-level need-supportive teaching. Reverse models tested whether T1 emotions predicted changes in motivation. Results Autonomous regulation predicted higher T2 enjoyment ( b = 0.21, p = 0.003) and lower boredom ( b = −0.13, p = 0 . 035). Need-supportive climate related to higher enjoyment and lower boredom and shame. Controlled regulation showed no unique associations. The association between autonomous motivation and enjoyment was stronger in secondary than elementary classes (interaction b = 0.31, p = 0.005). Reverse models showed no evidence that T1 emotions predicted changes in motivation. Discussion/conclusion Across short PE units, autonomous reasons for participation and need-supportive teaching appear to be linked to more favorable emotional trajectories. These findings support an affective-learning perspective in which students’ emotions are not merely outcomes of PE participation, but developmentally sensitive indicators of how motivational meaning is formed in instructional contexts.