Background Movement learning has been thoroughly investigated in the area of motor learning research. Although existing studies have contributed to a substantial understanding of motor learning, many have been criticized for their reliance on experimental designs where learning is decontextualized, simplified, and investigated in laboratory settings. Researchers have claimed that motor learning theories emanating from such studies are grounded on a dualistic approach to learning and that the theories are often difficult to apply in educational settings. More pedagogically-inspired studies of movement education have investigated movement learning, but the majority of this research has focused on teaching. This focus has left the process of learning somewhat unexplored. There is thus a need for empirical studies that investigate students' learning processes in educational contexts. Purpose The aim of this study is to explore, analyze, and understand how learners develop their movement capability when they are provided opportunities to choose different ways of learning activities. Theory and method We combine Ryle's and Polanyi's ideas concerning practical knowledge with Hirst's and Carlgren's idea of knowing as familiarity with a landscape. Ryle's notion of 'intelligent practice' is used in thinking of the kinds of actions individuals might engage in. Characterizing features of intelligent practice includes being sensitive to one's own actions, changing one's behavior as the result of mistakes, and profiting from the examples of others. We understand the development of movement capability as continuously expanding one's ability to discern nuances and their relationships. This perspective fits well with Polanyi's notion of focal and subsidiary awareness. Taken together Ryle's, Polanyi's, Hirst's, and Carlgren's notions related to knowing and learning inform our perspective on learning in movement education. Based on this perspective on knowing and learning, an action-oriented study was conducted. The researchers created pedagogical modules and collaborated with teachers and university educators to develop learning sequences in line with the needs of their respective groups. With each group, we produced data based on video and field notes. Three successful learners were chosen and followed in-depth with regard to their learning actions. Findings The findings show the learners' varying ways of exploring a movement landscape as playing around in the terrain; checking the map; investigating one chosen path; occupying the vantage point; imitating and actively observing. The findings suggest that oscillating between varying kinds of learning actions is an additional characterizing feature of 'intelligent practice.' Discussion The findings demonstrate how the learning of movement capability could occur when providing opportunities to engage in 'intelligent practice' while at the same time directing their focal awareness toward what is most beneficial to them. Opposed to a 'step-by-step' approach to learning, the learners come to know a movement landscape as extending one's capability to discern and differentiate details, nuances, and their relationships. The findings suggest that it may be beneficial for learners to get opportunities to oscillate between different kinds of learning actions.