In optics, real images can be projected onto a screen, while virtual ones always remain behind mirrors. This apparently straightforward distinction is based on complex premises which emerged in the Early Modern period, and its development went hand in hand with a transformation of the notion of image, which became detached from sensual perception. In this article I will outline this historical process, and argue that the distinction between a real and virtual image still implies a reference to visual perception which makes it problematic, yet useful for didactic purposes.

The notion of exploratory modeling constitutes a powerful heuristic tool for historical-epistemological analysis and especially for studying concept formation. I will show this by means of a case study from the history of particle physics: the formation of the concept of “strangeness” in the early 1950s at the interface of theory and experiment. Strangeness emerged from a broad space of possibilities opened up by exploratory modeling by authors working in communication and competition, and constructing both new questions and new answers. A systematic focus on exploratory modeling also helps compensate a bias towards the “right” developments still often present in historical investigations of theoretical work.