In science, Experimenter's regress refers to a loop of dependence between theory and evidence. In order to judge whether evidence is erroneous we must rely on theory-based expectations, and to judge the value of competing theories we rely on evidence, but to detect errors in experiments we must be aware of theoretical predictions, etc. Cognitive bias affects experiments, and experiments determine which theory is valid. This issue is particularly important in new fields of science where there is no community concensus regarding the relative values of various competing theories, and where sources of experimental error are not well known. What is science? There are different theories of what science is. ... Cognitive bias is any of a wide range of observer effects identified in cognitive science, including very basic statistical and memory errors that are common to all human beings (first identified by Amos Tversky and Daniel Kahneman) and drastically skew the reliability of anecdotal and legal evidence. ...

If Experimenter's regress acts a positive feedback system, it can be a source of pathological science. An experimenter's strong belief in a new theory produces confirmation bias, and any biased evidence they obtain then strengthens their belief in that particular theory. Neither individual researchers nor entire scientific communities are immune to this effect, see N-rays and Polywater. Pathological science is a term created by the Nobel Prize-winning chemist Irving Langmuir during a colloquium at The Knolls Research Laboratory, December 18, 1953. ... In statistical inference, confirmation bias is a type of cognitive bias towards confirmation of the hypothesis under study. ... The so-called N rays (or N-rays) were a phenomenon described by French scientist René-Prosper Blondlot but subsequently shown to be illusory. ... Polywater was a hypothetical polymerized form of water that was the subject of much scientific controversy during the late 1960s. ...