New computer model helps reveal how the brain both adapts and misfires(now.tufts.edu)
now.tufts.edu
New computer model helps reveal how the brain both adapts and misfires
https://now.tufts.edu/2025/10/16/flight-simulator-brain-reveals-how-we-learn-and-why-minds-sometimes-go-course
https://www.nature.com/articles/s41467-025-63994-y
20 comments
The “flight simulator” here is actually a software “brain simulator” built up from a model of actual brain cell biology. Like weather forecasting models, researchers can tweak various aspects of the brain and see what happens.
Headline is weirdly both dumbed down, and boosted with tech jargon.
As a reminder, you should check the original article (https://www.nature.com/articles/s41467-025-63994-y) to get more extensive and accurate information.
Indeed! Quite interesting.
title: The neural basis for uncertainty processing in hierarchical decision making
abstract: Hierarchical decisions in natural environments require processing uncertainty across multiple levels, but existing models struggle to explain how animals perform flexible, goal-directed behaviors under such conditions. Here we introduce CogLinks, biologically grounded neural architectures that combine corticostriatal circuits for reinforcement learning and frontal thalamocortical networks for executive control. Through mathematical analysis and targeted lesion, we show that these systems specialize in different forms of uncertainty, and their interaction supports hierarchical decisions by regulating efficient exploration, and strategy switching. We apply CogLinks to a computational psychiatry problem, linking neural dysfunction in schizophrenia to atypical reasoning patterns in decision making. Overall, CogLink fills an important gap in the computational landscape, providing a bridge from neural substrates to higher cognition.
title: The neural basis for uncertainty processing in hierarchical decision making
abstract: Hierarchical decisions in natural environments require processing uncertainty across multiple levels, but existing models struggle to explain how animals perform flexible, goal-directed behaviors under such conditions. Here we introduce CogLinks, biologically grounded neural architectures that combine corticostriatal circuits for reinforcement learning and frontal thalamocortical networks for executive control. Through mathematical analysis and targeted lesion, we show that these systems specialize in different forms of uncertainty, and their interaction supports hierarchical decisions by regulating efficient exploration, and strategy switching. We apply CogLinks to a computational psychiatry problem, linking neural dysfunction in schizophrenia to atypical reasoning patterns in decision making. Overall, CogLink fills an important gap in the computational landscape, providing a bridge from neural substrates to higher cognition.
For those who are frustrated with the lack of a fleshed out limitations section, it seems well covered under the peer review which is in the link above or directly below [0]
To me the paper is still very interesting, but concerns about computationally intractability and the hardness of approximation questions made me dig deeper.
> Specifically, our model aims to bridge biological circuits and computations of uncertainty in a tractable manner.
> To address this, we have carefully reframed our claims throughout the manuscript to emphasize that the model is a hypothesis generator rather than a definitive representation of biological circuitry.
Under the "All models are wrong, some are useful" I have no doubt this will be useful to some. But I will admit that their claims in the response to Reviewer Comment 4.5 that they "emphasize that the model is a hypothesis generator" doesn't match the published paper IMHO; and that negatively impacted my view of the claims in an admittedly probably unfair manner.
[0] https://static-content.springer.com/esm/art%3A10.1038%2Fs414...
To me the paper is still very interesting, but concerns about computationally intractability and the hardness of approximation questions made me dig deeper.
> Specifically, our model aims to bridge biological circuits and computations of uncertainty in a tractable manner.
> To address this, we have carefully reframed our claims throughout the manuscript to emphasize that the model is a hypothesis generator rather than a definitive representation of biological circuitry.
Under the "All models are wrong, some are useful" I have no doubt this will be useful to some. But I will admit that their claims in the response to Reviewer Comment 4.5 that they "emphasize that the model is a hypothesis generator" doesn't match the published paper IMHO; and that negatively impacted my view of the claims in an admittedly probably unfair manner.
[0] https://static-content.springer.com/esm/art%3A10.1038%2Fs414...
Thanks, we'll put that link in the toptext as well.
[stub for offtopicness]
Clicked the article because of "flight simulator" even though I should have stopped reading HN -->
Realized that the article has nothing to do with planes -->
Closed the article
Realized that the article has nothing to do with planes -->
Closed the article
It's actually an interesting news article, at least if you're interested in neuroscience. I can confirm, though, that it's nothing whatsoever to do with flight simulators, and have no idea why the author chose that particular simile.
"algorithmic psychiatry" yea, hard pass, but thank you
Oh ffs, stop trying to make science "relevant" by these lame titles. Science is cool, just do that.
"Please don't pick the most provocative thing in an article or post to complain about in the thread. Find something interesting to respond to instead."
https://news.ycombinator.com/newsguidelines.html
https://news.ycombinator.com/newsguidelines.html
Woe betide those who try to communicate science to a wider audience
While I agree with you in general, in this particular case the title is really very loosely related to the actual contents of the article.
I am 200% for communicating science. But while this is titled to attract young minds it's just too old vibed.
from chatgpt for this article:
“Inside the Matrix: How the Brain Rewrites Reality When Learning Goes Wrong”
“Neural Jedi Training: How the Brain Masters Skills—and Sometimes Turns to the Dark Side”
“Glitch in the Mind: Why Our Brain’s Learning System Sometimes Crashes”
“Braincraft: How We Level Up Knowledge—and Why We Sometimes Rage Quit”
“Neural Flight School: Why the Brain Sometimes Loses Control of the Plane”
“Mind Avengers: When Brain Circuits Team Up—and When They Fight Back”
“Stranger Thoughts: What Brain Simulators Reveal About Learning Gone Weird”
“Control+Alt+Brain: Rebooting How We Learn and Unlearn”
“Mind Simulator 2.0: Debugging the Brain’s Learning Software”
“Fast & Curious: How the Brain Learns at High Speed—Until It Skids Out”
“Inside the Matrix: How the Brain Rewrites Reality When Learning Goes Wrong”
“Neural Jedi Training: How the Brain Masters Skills—and Sometimes Turns to the Dark Side”
“Glitch in the Mind: Why Our Brain’s Learning System Sometimes Crashes”
“Braincraft: How We Level Up Knowledge—and Why We Sometimes Rage Quit”
“Neural Flight School: Why the Brain Sometimes Loses Control of the Plane”
“Mind Avengers: When Brain Circuits Team Up—and When They Fight Back”
“Stranger Thoughts: What Brain Simulators Reveal About Learning Gone Weird”
“Control+Alt+Brain: Rebooting How We Learn and Unlearn”
“Mind Simulator 2.0: Debugging the Brain’s Learning Software”
“Fast & Curious: How the Brain Learns at High Speed—Until It Skids Out”
Those all impose extraneous ideas on the article as well. "Rewrites reality", indeed. Why does it need to be a metaphor at all? Or a pun, ugh.
"A Cut Above - How the Prefrontal Cortex Dominates the Striatum Like a High-End Hair Salon"
"A Cut Above - How the Prefrontal Cortex Dominates the Striatum Like a High-End Hair Salon"
Sounds like the dude is tripping, though I will point out many profound ideas stumble out of the raw mind.
There is something about a hidden power in our minds to reinvent themselves into capable instruments of our own wills.
What these substance craving minds think of as reality is in actuality their thoughts feelings and beliefs.
Undeceiving the self rewrites reality with a sensibly competent version of one’s own self as an actor.
There is something about a hidden power in our minds to reinvent themselves into capable instruments of our own wills.
What these substance craving minds think of as reality is in actuality their thoughts feelings and beliefs.
Undeceiving the self rewrites reality with a sensibly competent version of one’s own self as an actor.
I don't believe you.
You prompted chatgpt to create movie based titles and then passed it off as regular output.
With the prompt "suggest 10 titles for this article" + the article text, I get the following _normal_ titles.
1. When the Brain Misreads the World: How Uncertainty Shapes Thought and Behavior
2. CogLinks: A Virtual Brain That Teaches Us How the Mind Adapts
3. The Neural Balancing Act: How the Brain Decides Under Uncertainty
4. Modeling Mental Flexibility: Simulating How the Brain Learns and Adapts
5. Inside the Decision Machine: How New Models Reveal the Brain’s Hidden Algorithms
6. Uncertainty, Meaning, and Misfires: Understanding the Neural Roots of Psychiatric Disorders
7. When Circuits Go Off Course: What a Virtual Brain Teaches Us About Mental Illness
8. The Thalamic Switchboard: Linking Flexibility and Habit in the Human Mind
9. From Neurons to Algorithms: Building a Bridge Between Brain Biology and Psychiatry
10. Toward Algorithmic Psychiatry: Simulating Brain Circuits to Decode Mental Disorders
You prompted chatgpt to create movie based titles and then passed it off as regular output.
With the prompt "suggest 10 titles for this article" + the article text, I get the following _normal_ titles.
1. When the Brain Misreads the World: How Uncertainty Shapes Thought and Behavior
2. CogLinks: A Virtual Brain That Teaches Us How the Mind Adapts
3. The Neural Balancing Act: How the Brain Decides Under Uncertainty
4. Modeling Mental Flexibility: Simulating How the Brain Learns and Adapts
5. Inside the Decision Machine: How New Models Reveal the Brain’s Hidden Algorithms
6. Uncertainty, Meaning, and Misfires: Understanding the Neural Roots of Psychiatric Disorders
7. When Circuits Go Off Course: What a Virtual Brain Teaches Us About Mental Illness
8. The Thalamic Switchboard: Linking Flexibility and Habit in the Human Mind
9. From Neurons to Algorithms: Building a Bridge Between Brain Biology and Psychiatry
10. Toward Algorithmic Psychiatry: Simulating Brain Circuits to Decode Mental Disorders