Neuroscience has a big problem with reproducibility and it only seems to be getting worse

Academic dishonesty will lead to end of the "scientific process" , and will herald a reversal in people moving away from religious dogma, and in this case the blame is entirely on the dishonest academics , the Arday case only magnifies this point (Cambridge dropped its "investigation" into the academic dishonesty behind his selection)

shocker

AI is going to utterly expose our ignorance of how the world works. We assume we have the right information for AI to produce real results, but I think it will reveal that our models of reality are just models and do not correspond to reality at all. We are living in a very different world than we believe.

I was mentioning how skeptical I am of neuroscience that causally links regions of the brain with behavior based on fMRI results, and the comment was received at the moment as if I was doubting something that was established beyond doubt. I feel vindicated. Ty.

This is a known issue in all life sciences. And discreetly shoved under the rug. There are objective reasons, mainly the primitivism of instruments available compared to the things to be investigated. In the end it all ends in poor statistical numerology based on dubious data.

A like for the fish behind the shoulder 😂

This idea gained speed when the Obama administration started The Brain Initiative program of 2013 . if you are in to brain play never pass up an opportunity to get grant money to MRI a fish . I would guess there's still money in the budget for this idea. The saga continues

They lack techniques to ethically isolate variables. Ambitious start, but not yet ready to play Carnegie Hall.

Reproducibility = Science

Criticism can be valid, but specialized fields require specialized knowledge. Otherwise, we risk moving from informed discussion to speculation.

It’s because there’s no singularity and no centre line. Reality breaks down because scientists look for answers that don’t exist; our existence creates the illusion of reality. If they try to find what makes reality work by breaking it down to its centre line, there’s nothing there and reality collapses. That’s why we’re almost at midnight on the Doomsday Clock: what you see is what you get—break it down and there’s nothing to see, There’s no point in the future where science becomes all knowing, because that’s impossible in our universe.

The nerve of them! I mean really......

Much of science seems to have this problem.

And what does this say about the validity of MRIs in general? Perhaps the study should be expanded.

Reliable, reproducible results are being obtained studying brain organoids, i.e. implementing a reductionist approach. There are even commercially available neurons-on-a-chip platforms, like those of FinalSpark.

Elon’s pet spews more propaganda.

Heh! My models produce a more accurate map of brain function than fMRI does. 🤣😉👍

What you put out is what you get back. It becomes a problem for you only if you define it that way. What you say, goes. Everywhere you are inside infinity, is the center. You are your own reality, therefore the one determining the spin, negative or positive. Our universe is holographic and every moment comes utterly neutral, servind double duty. Everything stems from what you choose to believe, first.

My salience network flagged this video as, well… salient.

Bad publishing incentives goes a very long way in explaining it

@dr_mcgilchrist ..can this affect your research?

@grok Yep, neuroscience is all fancy language and social democratic patronage politics

Excuses Excuses

Welll it shows me ONE thing people in general do not want criticism or critical thinking. Even in my field. everything has to be good.. but it isn’t … I am a the most critical cause that is the only way to define quality and value and and

It’s the problem of the null half section redux.

There are many reproducible methods in neuroscience, while fMRI sucks truly. Hard neuroscience goes well with electrophysiology

They suck, they really do. You’d have better results in the field by partnering electoral engineers & the guys responsible for understanding cpu substrate sublimation.

They talk a lot, image/measure very little

Thank you Bine, u r the real scientist. Love u.

@grok 如何理解

Basing a discipline on flawed research with inadequate sample sizes risks conclusions and predictions that are no more accurate than 'woo'!

Their ain't no neuroooooscience without no force of chaos stupids.

To human is error - even if you're a deterministic reductive physicalist nihilist hellbent on proving that you don't exist and that the universe does. @skdh why don't you ask me a direct question that you would want to know the answer to? You know I have answers.

Oh lord.