This paper aims to contribute to the growing field of research on teaching practices in higher education, with a focus on the prevalent yet unresearched practice of concept introduction in lectures. This small-scale case-study provides data which lend themselves to a comparative analysis, with a view to describing and accounting for the dispersion of practices which can be observed in spite of the similarity of teaching conditions: three teachers lecturing on the same formal definition of limits of sequences, in similar institutional conditions. A combination of an "a priori" mapped out space of possible actions, and concepts from the theory of teaching as decision-making allows for a fine-grained analysis of the data, consisting of video-recordings of the sessions, supplemented by pre- and post-interviews with the teachers. We identified two key factors -- knowledge of action-plans and 'subjective expected utility' (Schoenfeld, A. H. (2013). Classroom observation in theory and practice. "ZDM (Mathematics Education)," 45, 607-621) -- on the basis of which the instructors' engagement in teaching can be accounted for in a non-evaluative research perspective.