Common Logical Fallacies Committed by Scientists
The scientific method is the best system we have for discerning reality, but its mortal adherents are not immune from making mistakes. Though scientists do their best to shirk off the flawed modes of thinking that plague humanity, they are not perfect. They make errors just like the rest of us.
Such logical fallacies result in sloppy, substandard research, work that's stamped with the seal of science, but ultimately does little to elucidate the true nature of things.
In an effort to clamp down on the problem of poor logic in science, Espen A. Sjoberg, a PhD candidate at Oslo and Akershus University College, recently outlined a couple of the more common logical fallacies that scientists fall victim to. His helpful article is published in the journal Behavioral and Brain Functions.
One of the most prominent is the fallacy of "affirming the consequent," Sjoberg writes. It's the notion of seeking evidence to confirm hypotheses rather than searching for evidence to disprove them. The distinction seems subtle, but it can easily translate to false findings.
"As an example, imagine a hypothesis that states that patients with bipolar disorder have reduced cognitive processing speed, and we do a reaction time test to measure this. Thus, a fallacious line of reasoning would be: if bipolar patients have reduced cognitive processing speed, then we will observe slower reaction time on a test. We observe a slower reaction time, and therefore bipolar patients have reduced cognitive processing speed. This would be affirming the consequent, because the observed outcome is assumed to be the result of the mechanism outlined in the hypothesis, but we cannot with certainty say that this is true."
In fact, the observed outcome could be the result of an unforeseen factor unrelated to bipolar disorder. Evidence supporting a hypothesis does not singlehandedly prove it. You cannot completely confirm a hypothesis; you can only disprove it.
A second fallacy that arises is "false analogy," Sjoberg asserts. This is most prevalent in animal research. Essentially, scientists assume that because animals share many characteristics with humans, we can also assume they share other characteristics. For years, psychologists ran mice through mazes to glean insights into human psychology. That research paradigm is no longer utilized, as psychologists finally realized that just because mice and humans are both mammals with highly developed brains that share 97.5 percent of working DNA, that does not mean that mice think like we do. The same fallacious thinking plagues animal research into human diseases. Fewer than eight percent of positive cancer findings translate to humans. Less than one percent of stroke findings do.
Other errors in thinking also pervade modern science. Confirmation bias causes researchers to favor data that supports their hypotheses. Feeding into this is the notion of "sunk cost," essentially the fear of wasted time and effort. Some popular psychology theories persist on publication bias, the tendency to publish positive findings rather than null results.
"Thus, if we publish null results, it may seem that previous publications with significant findings were wasteful," Sjoberg says.
Considering the present state of the scientific enterprise, it will be extremely difficult to do away with the two primary fallacies that Sjoberg presented. To limit their problematic effects, he recommends a number of steps: Be aware of a study's limitations. Double-blind the experiment. Always strive to publish null results. And replicate, replicate, replicate.