Current Events > Two decades of Alzheimer's research may be based on deliberate fraud

(+) Yeah! (+)

I mean, from the outset though, there are serious issues with it. Like, only a very small percentage of people who get EBV go on to develop MS. So obviously something else is happening too, and it's that something else that we really care about.

I'd wait for systematic causal evidence, either in animal studies or human trials, before getting anywhere close to "we've identified the causal mechanism for MS".

The same type of thing happened in Physics (https://en.wikipedia.org/wiki/Sch%C3%B6n_scandal), and the peer review process was described as such:
"Referees need to assume that the authors are operating in good faith. Once you admit the possibility of fraud, it becomes nearly impossible for evaluate the paper.
It would be very damaging to the process of science if every paper had to be screened for fraud" -- Donald Levy, American Institute of Physics.

COVxy posted...

I mean, from the outset though, there are serious issues with it. Like, only a very small percentage of people who get EBV go on to develop MS. So obviously something else is happening too, and it's that something else that we really care about.

I'd wait for systematic causal evidence, either in animal studies or human trials, before getting anywhere close to "we've identified the causal mechanism for MS".

Yeah. I recall them talking about other conditions that are extremely common in MS patients like Vitamin D deficiency so the answer is probably more like you have EBV and 4 or 5 other things from this list that all push the immune system in this specific way that causes MS.

Well, shit.

The very nature of the academic circle-jerk is to blame. You have to be published to keep the grant gravy-train going, but you can't get published without asstons of references to other published papers.
The end result is articles that do what we all did in college: make shit up, put a long list of citations at the end, and just hope for at least a C (or rather, getting published in some journal) because nobody's really looking at the dogpile of papers too hard.

Questionmarktarius posted...

but you can't get published without asstons of references to other published papers.
The end result is articles that do what we all did in college: make shit up, put a long list of citations at the end, and just hope for at least a C (or rather, getting published in some journal) because nobody's really at the dogpile of papers too hard.

This isn't really accurate in my experience.

COVxy posted...

This isn't really accurate in my experience.

It's hyperbole than "accurate", yes.

Questionmarktarius posted...

It's hyperbole than "accurate", yes.

Having to cite work...doesn't pose any issues. A large part of an academic's job is keeping up with the literature, so while some papers may be miscited, it's not the general rule for certain.

Not only that, review processes aren't some sort of walk in the park.

I just don't identify what you wrote as a problem that exists.

COVxy posted...

I just don't identify what your wrote as a problem that exists.

Fair. The bigger issue to being published is the ludicrous paywall on both ends.
You gotta pay to get published, but also have to pay to have anything to cite - unless your institution is footing the journal subscription.

Questionmarktarius posted...

Fair. The bigger issue to being published is the ludicrous paywall on both ends.
You gotta pay to get published, but also have to pay to have anything to cite - unless your institution is footing the journal subscription.

If you're writing papers for academia you almost certainly get access to it as part of your benefits.

It's just easier for computer nerds, where "publishing" is making a project on Github or something, and "peer review" is the Issues tab.
Maybe actual academia needs to look at a sort of "open source" model.

I'm livid