Onpode
Cover art for AI predicts brain activity closely, but York researchers found it uses entirely different internal processes

AI predicts brain activity closely, but York researchers found it uses entirely different internal processes

November 22, 2026 · 7 min

Max Rivera & Clara Bennett

York University researchers tested AI vision models against primate neural recordings of 1,320 images and found a striking asymmetry: AI models predicted brain neuron responses fairly well, but brain activity could not equally predict the models' internal features — meaning matching outputs does not imply matching mechanisms.

Researchers at York University in Toronto — neuroscientist Kohitij Kar and postdoctoral fellow Sabine Muzellec — published a study in Nature Machine Intelligence challenging a foundational assumption in computational neuroscience: that an artificial neural network (ANN) predicting primate visual-neuron activity with high accuracy must be processing information in a brain-like way.

0:006:56
Get the next episode on Science →

Follow it free — new episodes land in your feed.

Or make your own — any topic, in minutes

More Onpode episodes on Science →

About this episode

AI vision models have gotten very good at predicting how neurons respond to images — good enough that researchers routinely treat high prediction scores as evidence that the models work like the brain. This episode examines why that reasoning has a hole in it, and what a team at York University did to expose it. The study, led by Kohitij Kar and postdoctoral fellow Sabine Muzellec and published in Nature Machine Intelligence, used a simple inversion. They showed primates 1,320 natural and synthetic images, recorded neural responses, and then asked the question almost no one asks: can the neural data predict the model's internal features? When you run that test between two biological brains, the prediction is symmetric — each recording predicts the other equally well. With AI models, it isn't. The models predict brain responses fairly well; the brain data does not predict the models' internals with the same fidelity. The episode is careful about what this does and doesn't mean. It doesn't overturn decades of work on neural network models of vision. Prediction remains useful. What fails is a narrower claim: that high predictive accuracy implies matching internal mechanisms. The episode also draws the line at overclaiming — this is visual cortex data, not a statement about consciousness or general intelligence. What you leave with is a concrete, practical test: two-way prediction plus mechanism-level comparison. And a genuinely open question about how much of the existing literature would survive it.

Frequently asked

Can AI models accurately predict brain activity?

AI vision models can predict primate neural responses fairly well. York University researchers showed this using 1,320 natural and synthetic images and recordings from primate neurons. However, high predictive accuracy does not mean the models process information the same way the brain does — the two systems can reach similar outputs via different internal routes.

What did the York University AI and brain study find?

York University researchers Kohitij Kar and Sabine Muzellec, publishing in Nature Machine Intelligence, found that while AI vision models predicted primate neuron responses well, the reverse was not true: neural data could not equally predict the models' internal features. Between two primate brains, the prediction was symmetric — the AI gap is not noise.

What is the difference between representational similarity and mechanistic similarity in brain-AI comparisons?

Representational similarity means two systems produce matching outputs or predictions. Mechanistic similarity means they use the same internal processes to get there. The York University study showed AI vision models achieve representational similarity with primate brains but fail mechanistic similarity — a distinction that invalidates the common claim that high prediction scores prove shared neural mechanisms.

Why is a two-way prediction test important for brain-AI research?

A two-way prediction test checks both directions: can the AI model predict neural responses, and can neural data predict the model's internal features? York University researchers found AI vision models pass the first test but fail the second. Between two primate brains, both directions are symmetric, so asymmetry with AI models signals a genuine mechanistic gap, not just noise.

Does the York AI-brain study mean AI doesn't think like humans?

The York University study's findings are specific to primate visual neurons and AI vision models — they say nothing about human cognition in general or consciousness. The research shows AI and primate brains differ in internal visual processing mechanisms, not that AI lacks human-like thinking broadly. That broader claim is not supported by this data.

Grounded in 7 sources
Deep Neural Networks and Brain Alignment: Brain Encoding and Decoding (Survey) ↗ · arxiv.org
Encoding neural representations of time-continuous stimulus-response transformations in the human brain with advanced deep neural networks | Imaging Neuroscience | MIT Press ↗ · direct.mit.edu
Decoding the brain: From neural representations to mechanistic models ↗ · doi.org
Convolutional Neural Networks as a Model of the Visual System: Past, Present, and Future ↗ · doi.org
Neural tuning and representational geometry ↗ · doi.org
From Brain to Machine: The Unexpected Journey of Neural Networks | Stanford HAI ↗ · hai.stanford.edu
York Study Finds Brain-Predictive Vision AI May Use Strategies… ↗ · letsdatascience.com
Read transcript

Max Rivera: Clara, hi, I have a confession before we start: I once trusted a friend's restaurant directions completely, and she'd never actually been there. She'd read a map. Totally confident, totally wrong about one closed road.

Clara Bennett: That's a good place to begin, because it's the same mistake a grad student could make in a vision lab. She picks the model with the best brain-prediction score and writes, 'the model captures how cortex processes faces.'

Max Rivera: Wait, that's the same thing? How?

Clara Bennett: Her model gives the right answers, like your friend gave the right directions. The question is whether it got them by the same route. What could go wrong with picking the top scorer?

Max Rivera: Huh. So, yeah, nothing looks wrong until the road's closed. The driver adapts, the map reader doesn't. And on normal days you can't tell them apart at all, right?

Clara Bennett: Right. Matching answers are called representational similarity. Matching internal routes are mechanistic similarity. Brain encoding and decoding models often treat predictive accuracy as a proxy for understanding the brain, and that's the shortcut.

Max Rivera: So the score is real. It just isn't the thing she wrote down.

Clara Bennett: That's why a team at York University didn't stop at the score. Kohitij Kar leads the study, and Sabine Muzellec, a postdoctoral fellow, is his co-author. They published in Nature Machine Intelligence. Their next move was to run the prediction backwards.

Max Rivera: Backwards how? Like, the neurons predict the model?

Clara Bennett: Yes. They showed primates 1,320 natural and synthetic images across ten categories, and recorded neurons. Then they asked two questions. Can the model's features predict the neurons? And can the neurons predict the model's features?

Max Rivera: And the answers match? I mean, I'd assume if A predicts B, B predicts A, but, yeah, I don't actually know that.

Clara Bennett: Between two brains, they do. Each recording predicted the other equally well. Symmetric. With the models, it wasn't. They predicted neural responses fairly well, but the neurons didn't predict the models' internal features equivalently.

Max Rivera: Wait, so the brains pass with each other? Then that's the ruler. The AI gap isn't just noise.

Clara Bennett: Exactly the point, and it means the test isn't impossibly strict. Biological systems clear it. The models are the ones that don't.

Max Rivera: Okay, the 300 stylized ones, the line drawings, outlines, artistic renderings. Is that where the cracks showed? Because a cartoon cat is exactly where I'd expect a model to fake it.

Clara Bennett: I can't tell you that. The stylized set was part of the design, but I don't have a result that says the asymmetry lived there, and I won't invent one. Also, I couldn't confirm the publication date, so don't call it fresh news.

Max Rivera: But hold on, because that's where the headline runs away with itself. "AI doesn't think like us." These are primate visual neurons. That's vision. That's a long way from thinking, yeah? Or consciousness, or whatever people picture.

Clara Bennett: Yes, and the study doesn't support that leap. The finding concerns primate visual neurons. It says nothing about human cognition in general, and nothing about consciousness.

Max Rivera: Because I think most people hear "brain-inspired" and they picture, like, a tiny brain inside the machine. And this knocks that over, but only a little.

Clara Bennett: Right. Now, the prior consensus was strong. Grace W. Lindsay's 2021 review in the Journal of Cognitive Neuroscience established CNNs as state-of-the-art models of biological vision. And Oota and colleagues' 2025 survey in Transactions on Machine Learning Research shows how routine it is now to use neural networks for brain encoding and decoding.

Max Rivera: So this is one study against a whole shelf of them. Does it refute that?

Clara Bennett: No. It's one data point. Prediction isn't suddenly worthless. What fails is a narrower claim: high prediction implies same mechanism. And "different mechanism" doesn't mean "inferior model." A model can be useful and still not be a copy.

Max Rivera: I saw some expert replies on X along those lines. That model-brain comparisons are approximations, not equivalences. And the generative-versus-discriminative thing is a false binary, so, I mean, if both ends of the slider are wrong, finding the middle isn't exactly shocking.

Clara Bennett: That's fair, and it's the right caution against overclaiming. The part that comes later is the grad student's sentence, and what a real test of mechanism would make her change in it.

Max Rivera: Okay, back to her draft, then. If the score is real but the sentence is too big, what does she actually type instead?

Clara Bennett: She crosses out 'captures how cortex processes faces.' She writes 'predicts responses.' That's all the score supports. Now, to keep the bigger sentence, she needs a second check: can the neural data predict the model's internal features?

Max Rivera: Huh. So she runs it backwards, like Kohitij Kar's team did.

Clara Bennett: Yes. In testing terms, it's a validity problem. The score is reliable. It just doesn't measure what she assumed it measured.

Max Rivera: Okay, but what would convince you the insides match? Like, I get the backwards check, but, I mean, if a listener asked me, 'what's the real test,' I'd, yeah, I'd stall.

Clara Bennett: Two-way prediction is the first filter. Then you compare internal structure directly. Nikolaus Kriegeskorte and Wei's 2021 Nature Reviews Neuroscience paper on neural tuning and representational geometry gives the framework for that. It asks whether the relationships among representations have the same shape in both systems.

Max Rivera: And if she skips all that, the failure is quiet, right? Some weird image she designs, the prediction misses, and she never traces it back to the model's insides.

Clara Bennett: Exactly that. So the takeaway is practical: validate brain-predictive models with two-way and mechanism-level tests, not prediction scores alone.

Max Rivera: Huh. I keep picturing my friend on the phone, giving me those directions so calmly. Nothing in her voice said she'd never driven it. And the score on the leaderboard sounds exactly that calm, yeah.

Clara Bennett: Right. Matching outputs is no longer enough evidence of matching mechanisms. And the open question is whether older results will survive a two-way test, with the neurons asked to predict the model too.

Max Rivera: So the road's still open for most of them, we just haven't driven it from the other side yet. Fair enough. I'd want to go check, I think.

AI predicts brain activity closely, but York researchers found it uses entirely different internal processes · Onpode