The transition from pre-scientific thinking to scientific thinking is characterised by the notion that knowledge is independent of the knower. What is true and right no longer was determined by kings and popes, but could be proven independently by anyone and for everyone to see - in principle. The process of producing knowledge became transparent and the knowledge itself accessible and shared, in contrast to the secrets of alchemy.
The chapter on scientific thinking, aims to explore what doing science means, how this has changed over the course of history and how we might assess the rigour and quality of different ways of knowing.
It starts with some historical examples of forms of inquiry. For example, we talk about Jon Snow, a British physician who traced the source of a Cholera Outbreak in SoHo, London in 1854. In what we nowadays would call Data Science, he systematically mapped out patterns of infections and thereby correctly identified a water pump as the origin. Doing so, he also proved that cholera was in fact transmitted via water and not, as believed at the time, by particles in the air (Miasma theory). Other examples include the famous experiment by Galileo, dropping two spheres of different mass from the leaning tower in Pisa to prove that they descended in the same time (an experiment that was replicated with a feather and a hammer by the Apollo 15 crew on the moon). Closer to Computer Science, we speak about Fitts Law, Alan Touring’s theoretical work and a controversial Facebook study that showed evidence for emotional contagion through social networks.
Looking across these examples we derive what steps characterise the classical empirical scientific method: it is about observing a phenomenon, speculating about any causal relationships, testing these theories and formulating more general knowledge (observation, induction, deduction, testing, evaluation).
We then take a detour into historical philosophy of science. We briefly touch on some key thinkers and their main theses that helped to bring about and foster the scientific revolution (and with it, modernity and fundamental changes in our societies): Nicolaus Copernicus (helio-centric world view), Francis Bacon (knowledge is power), René Descartes (cogito ergo sum - the Matrix), John Locke (empiricism) and Immanuel Kant (idealism).
Reflecting on the impacts that this rational and modern worldview still has on our world, we discuss technological utopias (“there is an app for that…”) and dystopias (1984, Edward Snowden).
In the remainder of the first of the two units in which scientific thinking is presented in class, we define fundamental terms in science: hypothesis (and what are good and bad ones), falsification, experiment, (in)dependent variables, method, the difference between correlation and causality, and the predictive quality of theory. With each of these terms we seek to provide examples from research that is conducted at our faculty.
The second unit on scientific thinking starts with a short recap of the above and then turns to the concept of science paradigms. We discuss Thomas’ Kuhn concept of scientific revolutions and paradigm cycles.
As an example, we step through the three major shifts in Human-Computer Interaction - the “three waves of HCI” - from the human factors perspective, to cognitive models, to a situated and embodied understanding of interaction.
Using this example, we more generally talk about three very different science paradigms that exist in computer science: (Post-)Positivism, Constructivism and Critical Theory. We look at examples of studies within Computer Science that would argue their insights from within these paradigms. We turn to the question about how to identify “good” research within each of these paradigms (e.g. internal/external validity, generalisability, replicability versus relevance, credibility, authenticity, transferability).
In a small thought experiment, we discuss how different science paradigms would allow researchers to ask different questions and inquire into different problems in a particular context, an airplane cockpit.
Also, to show that the philosophy of science is still moving and still grapples with the fundamental ontological and epistemological questions on what there is and how we can know about it, we briefly discuss one example of a more contemporary approach, the Agential Realism by Karen Barad.
We end this chapter by talking about science a social activity with many constraints other collecting data and constructing knowledge: we speak about peer-review, the publishing culture and the pressure to publish in pursuing an academic career. This leads us to debating “fake” science (automatic paper writing algorithms), the publishing industry, the replication crisis and the systematic incentives (or rather, disincentives) for academics to produce good science.
Next: mathematical thinking
Calls for discussion
Where do you think we could improve this chapter? Are we missing essential bits?
We always appreciate ideas for exercises that can help students comprehend scientific thinking core concepts.
Do you know empirical or experimental scientific projects from informatics that can be used as examples?