Friction Theory — in plain English

Tomas Pødenphant Lund · Independent researcher, Aarhus

I almost stumbled into studying language models, the kind of artificial intelligence ChatGPT is built from. I wanted to see one concrete thing: could I make them better by letting them doubt, feel their own uncertainty, before they answered? What I found was something else, and bigger. Across area after area where we assume there's a deep difference between us and the machine, the difference actually isn't that large. We can recognise ourselves in it. But the difference that does remain, that one is beautiful, and it matters. This page is about both: how little really separates us, and what it is that does. The patterns reach further than the brain, but the human end is the easiest to recognise, so that's where we begin.

Three places to start

Start here
What is a race? Explained with water The whole theory rests on one picture: water running through pipes, and the tank that fills first wins. Three minutes with simple drawings, and the rest of the site makes sense.
Understand the mechanics →
Learning
The physics of learning Three of the most replicated principles in learning research, cognitive load, desirable difficulties and safety-as-prerequisite, never talk to each other. They follow from the same constraint: how much can be held open at once while something has to be chosen.
Memory page →
The mirror
The less you know, the surer you get We are most certain right where we know least. That curve turns up in a language model while it learns, and a machine has no ego to defend. Measured directly in four models.
See the curve →
Using AI well, in practice

A practical, evidence-based guide grounded in this research: AI is the engine, you are the pilot. How to feed a model, check it against an answer key, and get it to think in new directions.

Behaviour design, from one mechanism

Why the classic nudges (social proof, defaults, scarcity) work, and how to pick the right one: find which of four fields is blocking, then use the technique that acts on it. Behaviour design: find the field that blocks.

"It's just probability" — for the skeptics

The answer to the objection that we can't trust language models because they only work on probability: so does the brain you are comparing it to. It’s just probability (or the plain version).

The rest of this page goes deeper. If you want a one-paragraph version: every system that has to choose under finite time and resources runs an internal race between alternatives, one wins, the others are suppressed at a cost. That cost is what I call friction, and the same structure shows up at every scale that's been examined.

What I'm actually studying

Every thought you have, every feeling, every movement of your hand: they are all expressions of computations your body performs.

That sounds odd put that way. "Computation" sounds cold and deliberate, like a calculator running, like someone sitting inside you working out the answer. That isn't what is happening. What is actually happening is closer to a sequence of probability outcomes.

A small example. Inside one of your cells, a stretch of DNA needs to pick up one starting molecule or another. Which one it picks up depends on what is floating nearby, and in what concentration. There is no decider. There is a distribution of molecules and a probability that one lands before another. The cell then produces whichever protein follows from that landing. Multiply that across millions of events per second across your body, and you have something that "computes" continuously without anyone running the show.

The same picture scales up. When you decide between two options, you are not running a deterministic calculation. You are letting competing candidate-answers race each other in your brain under conditions of finite time and finite resources. One of them wins. You experience the winner as "I chose X." You do not experience the race.

How learning actually works

Imagine dragging your finger through a thin layer of water on a tiled bathroom floor. The water shifts; a channel forms. Drag your finger through the same path again, and it's a fraction easier, because the channel is already there. Physicists call this hysteresis: the system carries traces of its own history.

Your brain works the same way. Routes you use a lot leave traces, and the traces make those routes more probable next time. That is what learning is, at the substrate level: not magic, not a uniquely biological mystery, just probability shifted through trace-accumulation.

This is what Friction Theory is about. The price the system pays every time it has to resolve competing candidates into one committed outcome (in time, energy, information) is the friction. Friction Theory is the formal name for the framework. Behavioural Friction Theory (BFT) is its biological version, applied to nervous systems and organised around four computational fields (Safety, Meaning, Capability, Effort) and five regulatory layers.

Why information-dumping doesn't teach anyone, anything

Even language models, which are literal computers designed to absorb information, cannot be taught by being shown more of it. Paper 2B shows this directly. Throw information at a model and you get a model that hallucinates confidently about whatever you threw at it.

If language models, designed to be teachable, cannot be taught by information-dumping, why do we assume humans can be? We have been thinking about it as motivation. It is physics. You do not learn information; you learn the trace information leaves.

The full version of that argument (what it means for how we teach, and why Bjork's "desirable difficulties" is physics rather than a pedagogical choice) is on the Memory page (plain) or the Learning page (more academic).

What I do, day to day

I study language models in order to understand humans, which is the opposite direction from what most people expect. The conventional view is that language models try to mimic human language, so studying them tells you about the mimic, not about humans themselves. I am arguing the opposite.

Many of the things we thought were uniquely human turn out to appear in language models too. The reason isn't that the models were trained to imitate human cognition. The same race runs in both: parallel candidates racing under finite resources, with one winning. That same race produces the same characteristic behaviours:

→ The full tour: LLMs aren't calculators — a dedicated page on all the surprising things about language models. All the things you would expect to be typically human.

When language models reproduce these phenomena, they are not faking. They are telling us the phenomena are not specifically biological. They are structural consequences of any architecture that lets competing options race toward a deadline, with only one winner. The brain is one such system. A transformer is another. They look alike where their architecture is alike, and diverge where their architecture diverges.

What humans have that language models do not

The architecture is shared, but that isn't the whole story, and this is where it turns beautiful. The places where we and the language models differ are the most important of all. Because when so much of what we thought was distinctly human turns out to be something we share with a machine, what remains, the part the machine does not have, stands out all the more sharply. That is where the human sits.

They do not have loss aversion: the human tendency to fear losses about twice as much as we enjoy equivalent gains. This is because loss aversion fundamentally comes from the fact that you can die if you choose wrong too often. Humans, mice, and bees all have mortality; language models do not. This is a new explanation for an old finding: loss aversion is not a universal cognitive law. It is a consequence of having a body that can die.

They do get surprised: words they did not see coming draw measurably more attention, and breaking the format makes them measurably resist. Both are measured directly. The still-open question is not whether they react, but whether they can tell the two apart: whether there is a layer that distinguishes "new information I should update on" from "something I want to push back against". That layer has not yet been seen in a language model.

They do not have memory between conversations. When you start a new conversation, the language model starts over. This means a huge area of human cognition (from Ebbinghaus's forgetting curve to spaced repetition) literally cannot be tested on language models, because they lack the basic substrate for it.

Could they doubt? Yes

The question I started with, whether a model can feel its own uncertainty, turned out to have a concrete answer. There is already a small signal in any language model's output that tells you when it is torn between several possible answers (I call it Competing Routes). Ask the model to reconsider when the signal is high, and let it hold back from answering when the signal is very high, and it lifts performance noticeably, with no retraining and no extra computer. That is the practical end of all this. The full explanation is on the Paper 3 page.

Where humans and language models meet, and where they diverge

Short version:

They meet on: anchoring, hysteresis, confirmation bias, mode-shift cost (the cost of switching between modes of thinking), expertise reversal effect (instructions that help beginners hurt experts), surprise-weighted encoding, and many classical cognitive biases.

They diverge on: loss aversion (requires mortality), spaced repetition (requires between-session memory), and field-organised friction (Safety / Meaning / Capability / Effort, specific to organisms that can die, move, and consume metabolic energy).

The pattern shows up almost everywhere

Once you start looking for the race-architecture signature, you see it at scales that have nothing to do with brains or computers. The same characteristic curve (performance peaks in the middle, drops off at both ends) appears in:

An optimum under limited resources how well it goes a quantum particle a chemical reaction a faint signal stress learning fractions of a second a whole lifetime Too little: nothing happens. Too much: it breaks down. In between it works.
Five of the seven, drawn side by side. The time scale runs from fractions of a second to a whole lifetime, and the shape holds all the way. These are not five analogies; it is the same constraint read off in five places.

Seven different things, spread across roughly forty orders of magnitude in time. They all show the same shape because they all face the same constraint: multiple options have to resolve into one outcome under finite resources. Paper 10 walks through the seven phenomena and the race-framework behind them.

The suggestive implication: biological cognition and physical systems may share the same race-structure, differing in substrate not in shape. Humans hit a performance peak in the middle of the challenge range; qubits hit one between coherence and decoherence; the proposal is that both can be read under the same race-vocabulary, with the substrate differing and the constraint shared. Behavioural patterns we usually think of as "psychological" (the inverted U, hysteresis, anchoring, mode-shift cost) may turn out to be describable in the same race-vocabulary as the physical ones. This is a shared lens, not a claim that the substrates are identical.

That is the program. The papers develop the formal apparatus, the empirical signatures, the testable predictions, and the falsification criteria.

My papers

All papers are open-access preprints on Zenodo:

The original version of the theory, focused on biological systems. 21 testable propositions, four functional fields (Safety, Meaning, Ability, Effort), and one mechanism — the RACE model — that ties it together. Just updated to version 7.
Read on Zenodo · Technical page
The instrument in the programme. In a language model the competition between answers can be read directly, token by token, with no model fitted to it. Everywhere else that quantity has to be inferred backwards from an outcome. The paper is cut to that measurement and keeps a ledger of its own signatures: two meet the standard, one is withdrawn. The principle itself, that choosing costs something, belongs to others and the paper credits it.
Read on Zenodo · Technical page
The practical paper. Strategy pipeline alone +7.7 to +20.8 pp; combined with calibrated abstention reaches +12 to +21 pp on the four cells where both were measured. On SimpleQA, the combined pipeline lifts Qwen3-235B past GPT-4o and GPT-4.1. Calibration costs about $1.50 per setup.
Read on Zenodo · Technical page
Why do fine-tuned language models give more confident wrong answers than RAG on the same knowledge? Every gradient step strengthens the route that won and presses the alternatives further down; after thousands of steps they are gone. It reads token by token as the distance to the second-best answer: 5.46 nats under ICL, 17.85 after 30 epochs. Variation in the training data measurably slows it. And fine-tuning left the model worse at applying knowledge than if it had never been given it.
Read on Zenodo · Technical page
For teachers, instructional designers, communicators, and anyone who has tried to explain something complex to someone else. Learning science has three big traditions (cognitive load theory, desirable difficulties, psychological safety) that almost never talk to each other. This paper argues they are all describing the same underlying physics from three different angles. From four substrate-level concepts (race-architecture, friction, hysteresis, Net Friction Rule) five classical learning findings fall out as derived consequences. Three classroom failure modes (dump, dilute, ambiguity-without-commit) get diagnostic recipes; four falsification conditions specified.
Read on Zenodo · Technical page
We try to produce behaviour by giving people more information: one more policy, one more course, one more warning. But compliance is something you do, not something you know. More information rarely changes what people actually do. The paper explains why, through friction, and what it takes instead.
Read on Zenodo · Technical page
We measured the Dunning-Kruger effect directly inside a language model: most confident exactly where it knows least. The confidence tracks how the competing routes resolve, not real competence, which turns the model's own logprobs into a measurement instrument for human-style overconfidence.
Read on Zenodo · Technical page
You can install a value field's function (such as Safety) into a model by fine-tuning on the function, not on facts, and then watch it change what the model attends to. The field-installable model becomes a controllable model organism for human cognition, the constructive counterpart to using an AI as a subtraction control.
Read on Zenodo · Technical page

In progress, not yet out

Several papers are in preparation and will be released in the coming weeks:

Contact

Email: tomas.lund@frictiontheory.org
ORCID: 0009-0000-4724-2427
LinkedIn: linkedin.com/in/tomasplund