The Atlantic magazine recently published another article about the replication crisis, focusing on why disproven research is still studied. The fields of science are littered with what Steven Poole refers to as "intellectual zombies," disproven ideas that refuse to die.
An illuminating example are 18 candidate genes that are ostensibly linked to depression, which upon a second, more thorough look may have nothing to do with depression at all. These genes have been heavily researched, all based on foundation of research that turned out to be nothing more than sand.
Graduate students are uniquely harmed by these unreplicatable experiments. While there are real fiscal and societal costs, there are also unseen personal costs. The “waste of 1000 studies” is a waste of many more years that graduate students spent trying to make these experiments work.
We have five or so years to learn and work on a project. And critically, we often have five years to publish said project. If you’re starting a project on a foundation that doesn’t exist, you’re set up for failure. And in cases like the one the article discusses, where the evidence seems strong, that failure feels so personal. The logical assumption is that it’s your fault, as the lowly graduate student.
It can get worse when the thing you’re seeking to build off and replicate is from past research in your own lab. Poor mentorship might mean you’re not just unable to fight the zombie but are also forced to reanimate it.
It doesn’t help that research is a job that relies on zeal and curiosity, which are often fragile things. Intellectual zombies might not eat your brain but they eat away at your faith in the system.