2 ms·
forensic image analysis is cool and all, but i think maybe instead of: "A neuroscience image sleuth finds signs of fabrication in scores of Alzheimer’s article
by a-dub 4y ago
forensic image analysis is cool and all, but i think maybe instead of:
"A neuroscience image sleuth finds signs of fabrication in scores of Alzheimer’s articles, threatening a reigning theory of the disease."
it would be better to see:
"A neuroscientist conducts an irrefutably strong study that upends 16 years of theory on the disease process behind Alzheimer's."
i don't care if someone cheated, i care that their results are discredited by scientific processes, not forensic review.
- appleflaxen 4y agoThe scientific process is the systematic identification of what is true. Wikipedia: "the scientific method involves careful observation, applying rigorous skepticism about what is observed, given that cognitive assumptions can distort how one interprets the observation." So identifying fraud is the scientific process.
- a-dub 4y agosure. but i think i'd rather see biologists debunk prevailing theories in biology by doing biology while in the process getting better at biology rather than dabbling in forensic image analysis. (although, who knows, maybe said dabbling could lead to some new meaningful insights down the line) i guess the bigger point is that the act of topical inquiry should have inbuilt mechanisms for discarding ideas and approaches that turn out to be dead ends, without having to rely on them actually being fraudulent (they can be, and often are, completely legitimate, yet also completely wrong). we shouldn't be discovering bad ideas in science by fraud detection, we should be discovering them by mainstream scientific process.