Executive functions, also known as cognitive self-regulation, are important for young learners in the classroom as they support school readiness, academic achievement and social competence (Blair & Razza, 2007; Riggs et al., 2006). Children show individual differences on these skills and researchers aim to understand how these skills can be enhanced. School offers an opportunity to introduce change. However, it is not yet clear how children’s executive functions can be best promoted in a school context. Many interventions that have been examined for their effectiveness included multiple components, making it difficult to discern the active ingredients. Effective curricula have in common that children are active and engage in hands-on learning where they have agency and choice (e.g., Diamond & Lee, 2011; Lillard & Else-Quest, 2006). This PhD study focused on N = 218 typically developing 4-5-year-old children’s executive function development in N = 32 low- to middle-income classrooms in England and aimed to explore how children’s daily experiences in these classrooms relate to their executive function development. Findings showed that children’s proportion of time spent in teacher-led activities, child-led activities and transitions was not related to children’s executive function development. Also, children who started the school year with lower executive functions did not benefit differently from proportion of time spent in teacher-led or child-led activities. Furthermore, no specific association between the proportion of time spent in child-led activities and children’s self-directed executive functions (measured with tasks where children guide their own behaviour) was found. Findings also did not show an association between the proportion of time spent in teacher-led activities and children’s externally-driven executive functions (measured with tasks where children follow instructions). Findings indicated that children who spent more time in less purposeful activities during child-led time had lower executive functions at the end of the school year. Furthermore, the observed level of instruction generally focused mostly on basic skills instruction. Teachers provided the highest levels of instruction in spontaneous interactions with children during free play. Taken together, the classroom characteristics that were assessed seemed to play a minor role in children’s executive function development. In turn, children’s executive functions at the beginning of the school year and their concurrent vocabulary skills were consistently related to their performance on executive function tasks at the end of the school year. The second part of the PhD study focused on different measurement approaches of children’s self-regulation and executive functions. Children’s self-regulation and executive functions were measured with lab-based assessments, teacher reports, and child observations. It was tested if children’s self-regulation and executive functions could be categorized by levels of ecological validity and child agency. Findings showed that children’s self-regulation and executive functions could be categorized by high and low ecological validity. However, results indicated that children did not perform differently on tasks and activities of high and low child agency. First, this study adds to the understanding of how executive functions can be enhanced in the classroom. My findings show that providing more time in child-led activities might not be one of the key ingredients. Other potentially important factors such as the role of the teacher are discussed. Second, children used their executive functions more similarly in naturalistic compared to lab-like contexts. This is consistent with previous work and indicates that it is important to measure children’s executive functions with ecologically valid measures if inferences to children’s behaviour in the classroom are intended.