English: Kevin Dunbar and Jonathan Fugelsang describe how luck can play a particular role in certain stages of the scientific method.

With careful controls, anomalies that may not fit existing theory are more easily identified. At first scientists tend to blame human error: they revise, relatively without assistance, their methods. If the number of unexpected findings rise and reach a threshold, the existing theory no longer seems adequate. At this point scientists tend to collaborate more with thinkers who can bring ideas or analogies from other specialities.

This figure uses two images from WikiCommons: File:BLUELINE.jpg, and File:Beauty Girl Surprise.jpg

Otherwise this figure is entirely my own work, based on the ideas of Dunbar and Fugelsang in the following paper:

Dunbar, K., & Fugelsang, J. (2005). Causal thinking in science: How scientists and students interpret the unexpected. In M. E. Gorman, R. D. Tweney, D. Gooding & A. Kincannon (Eds.), Scientific and Technical Thinking (pp. 57-79). Mahwah, NJ: Lawrence Erlbaum Associates.