Konrad Kording

Here’s why more than 50,000 psychology studies are about to have PubPeer entries PubPeer will see a surge of more than 50,000 entries for psychology studies in the next few weeks as part of an initiative that aims to identify statistical mistakes in academic literature. The det…

statcheck has been posting automatically on pubpeer: retractionwatch.com/2016/09/02/h... -- there is a lot more coming our way. What do we want it to de?

But if we want it or not, the rules and their enforcement will change very soon. So let’s work on them being a little bit less stupid.

Understand which (soft) rules are the most useful. This is crucial right now because the set of rules can be changed from tiny problematic set (because human io/memory) to much bigger problematic set. Sometimes more is different. Sometimes more is stupid.

Can you read the op and see that it has a “it’s possible but there are downsides, let’s think carefully” structure?

Where did I say what I want to do? I implied we likely will do it (and already do it), pondering the tradeoffs.

I gave it as an example of a rule we do enforce. This one has a lot of bad(but also some good)

Turtles all the way down?

What about “your p value must be smaller than 0.05”?

I presume without rules for biomedical science our healthcare would be much worse. Like “don’t sell drugs to masses without an rct“ , you need to share your code and data, etc. almost all of them have exceptions. Plenty apply to much of science.

Obviously not. But also, obviously, your papers would be among the worst to automate review of.

The review process will be automated in significant parts. It probably is (illegally) automated by 15%-50% at least at this point of time

Here is the course: jrudoler-teaching.github.io/understandin..., and the github github.com/jrudoler-tea...

Its a new term and a lot of us are preparing their lectures. I am a huge fan of online ways of teaching. And my student Joey Rudoler @jrudoler.bsky.social made a great online stats course. Shiny app and all. Modern take on stats. Thought some of you may appreciate.

We can now define a set of rules of good science and automatically flag violations, and help users avoid violations. But we may kill creativity in the process. Which rules should we enshrine? How should we manage a system where some need rules and others are stifled by them?

There are profound insights in the simple. And simple explanations for the profound. Russian stacking babushka dolls all the way down.

I am old enough that a 20% forgetfulness rate strikes me as an obvious property of life.

Can you have different thoughts in the same world? The results will shock you ;)

four brain maps showing noise ceiling estimates in response to image presentation

New CNeuroMod-THINGS open-access fMRI dataset: 4 participants · ~4 000 images (720 categories) each shown 3× (12k trials per subject)· individual functional localizers & NSD-inspired QC . Preprint: arxiv.org/abs/2507.09024 Congrats Marie St-Laurent and @martinhebart.bsky.social !!

Is involuntary hospitalization killing people instead of saving them? www.newyorkfed.org/medialibrary...

a bald man is sitting in a chair and pointing at the camera ALT: a bald man is sitting in a chair and pointing at the camera

I'm curious if any senior nonhuman primate Neuro-AI researchers would be interested in joining Queen's University if we were to obtain a research chair position (full professor level)?

Could you please send me a confidential message to indicate your interest? Gunnar.blohm@queensu.ca

Dl

Neuromatch do and neuroAI courses contain a lot.

This line strikes me most.

We're doing this work not for our blink-of-an-eye lives. It's for our whole species. What future generations build from depends on our integrity in the present.

It's wrong to corrupt the body of knowledge humanity has been building since its origins for personal gain.

Now pain maybe different from more “emotions” but I totally think all of them are results of inference and can be normative modeled.

Pain: A Statistical Account Perception is seen as a process that utilises partial and noisy information to construct a coherent understanding of the world. Here we argue that the experience of pain is no different; it is based o...

We have to imho journals.plos.org/ploscompbiol...

Sensory responses of visual cortical neurons are not prediction errors Predictive coding is theorized to be a ubiquitous cortical process to explain sensory responses. It asserts that the brain continuously predicts sensory information and imposes those predictions on lo...

1/3) This may be a very important paper, it suggests that there are no prediction error encoding neurons in sensory areas of cortex:

www.biorxiv.org/content/10.1...

I personally am a big fan of the idea that cortical regions (allo and neo) are doing sequence prediction.

But...

🧠📈 🧪

This is a great paper that I missed at the time. Worked example of how neuroscience methods could lead to a wrong conclusion on figuring out a tiny neural network. Like @kordinglab.bsky.social "could a neuroscientist understand a microprocessor" but even simpler.

I’”ll answer with a blog post ;)

Thanks! I missed a key thing!

The thing that drives me is just that attractors are now part of a hype cycle, with most papers claiming mechanistic insight with methods that largely does not provide that. In a manner quite similar to those of tuning curve research in the past.