Science is often thought to be self-regulating, in the sense that faulty scientific theories would easily get refuted and filtered out, while correct ones would survive the tooth of time and eventually get accepted by the scientific community. This presumed process underlies the worldwide confidence that science has brought us closer to an accurate understanding of nature and the universe, with scientific consensus by the majority of experts as a reliable foundation of scientific education and various policies on which our daily lives depend. Also as scientists and engineers, we often need to rely on the good judgement of our colleagues from the past and in the present, through institutionalized aids like conferences, peer-review and the scientific method. Yet, how much can we truly trust this self-regulating mechanism of science? And how much does this affect our field of plasma physics and chemistry? In this oral talk, I will present my personal perspective on this matter by means of a few examples. Each of these examples is taken from my personal research experience in what I call the plasma physics of liquids [1], i.e. the combined study of plasma-liquid interaction and in-liquid plasma generation, as well as their applications. They illustrate a tendency of scientists to favour theory over experiment, even in the presence of empirical data contradicting the theory. Additionally, they demonstrate how scientists, when confronted with multiple competing theories, often gravitate toward the theory closest to their worldview, even when colleagues in a neighbouring field prefer one of the alternatives. In conclusion, science can be self-regulating if conflicting views are forced to meet, but progress may just as well be delayed, perhaps indefinitely, if such confrontations remain ignored, underrepresented or unpursued.