LESSWRONG
LW

Johannes C. Mayer — LessWrong

No, but great idea! I'll likely run one. I already ordered some microcristaline cellulose and designed an experimental protocol.

Replying toGlucose Supplementation for Sustained Stimulant Cognition

Johannes C. Mayer2mo*

Glucose Supplementation for Sustained Stimulant Cognition

Glucose monohydrate powder, and then put it in capsules. (Dextrose/D-glucose monohydrate to be extra precise.)

Replying toGlucose Supplementation for Sustained Stimulant Cognition

Johannes C. Mayer2mo

Glucose Supplementation for Sustained Stimulant Cognition

It's possible I have a metabolic disorder that wouldn't be detected by regular blood tests. And yes the amount of glucose is absolutely tiny. I also bought a blood glucose meter. It doesn't show elevated values at all from supplementing glucose. When eating it does increase measurably in line with what is normal. I do have sleep apnea which might do weird stuff, like give you diabetes. I do have a CPAP though maybe there is still some effect from that.

I don't quite understand why it works. But it seems really strong of an effect. Once I increased the amount of MPH I took, then I took 0.6g of glucose, and it suddenly made me feel a pressure in my heart. The effect of the MPH was now too much. Something was throttling the effect of MPH before taking the glucose. And somehow taking the glucose stopped the throttling. This happened in less that 10 minutes. Probably less than 5.

Glucose Supplementation for Sustained Stimulant Cognition

Johannes C. Mayer

2mo

Observation

I take 60mg methylphenidate daily. Despite this, I often become exhausted and need to nap.

Taking small amounts of pure glucose (150-300mg every 20-60 minutes) eliminates this fatigue. This works even when I already eat carbohydrates. E.g. 120g of oats in the morning don't prevent the exhaustion.

Proposed Mechanism

Facts:

Wiehler et al. (2022) found that cognitive fatigue correlates with glutamate accumulation in the prefrontal cortex.
Glutamate is the brain's main excitatory neurotransmitter.
Excess glutamate is neurotoxic.

Hypothesis-1: The brain throttles cognitive effort when too much glutamate has accumulated.

Facts:

Glutamate is cleared by astrocytes.
This process costs 2 ATP per glutamate molecule (Escartin et al. 2006).
The ATP comes from astrocyte glycogen stores.
Sickmann et al. (2009) found that blocking astrocyte glycogenolysis impaired

Johannes C. Mayer3mo

Drugs Aren't A Moral Category

In the past I would have said when ask: "obviously not all drugs are bad" without being pressed. But when it comes to moment to moment decision making I would have subconsciously weight things such that not taking drugs is better. That is the pernicious thing. It's not about what people say when pressed. It's about how they make decisions moment to moment.

It seems that I used moral strictures subconsciously, while my stated position was almost be the opposite of these strictures. And both are---like you said---didn't really make sense.

Drugs Aren't A Moral Category

Johannes C. Mayer

3mo

Are drugs good?

This question doesn't really make sense. Yet Western society answers with a firm "NO".

I have ADHD and have a prescription for Methylphenidate (MPH). Often I don't feel like taking it. Shouldn't I be able to just do the right things? I can just decide to be productive. Right? Needing a drug to function feels like a flaw in who I am.

I also have sleep apnea. That means I stop breathing at night, which makes the CO2 concentration in my blood rise until I wake up. This has quite bad health implications if untreated. However, I have a CPAP which is a device that constantly blows air in my mouth during... (read 472 more words →)

Replying toDominance: The Standard Everyday Solution To Akrasia

Johannes C. Mayer3mo

Dominance: The Standard Everyday Solution To Akrasia

I still have a strong dislike of mathematics which I acquired by doing the first semester of a German mathematics degree. I think doing that was actively harmful in certain ways. Not sure if it was net negative though. A similar thing happened when studying GOFAI and Chemistry at university.

Each time that I tried to do some accountability buddy thing it completely failed.

The things that actually worked for me is taking methylphenidate, and seriously trying to answer the question of what I think the best thing to do is (in the moment). Once I figure out the thing that I think is actually the best thing to do, it becomes easy to... (read more)

Evaluation Avoidance: How Humans and AIs Hack Reward by Disabling Evaluation Instead of Gaming Metrics

Johannes C. Mayer

3mo

Have you ever binge-watched a TV series? Binge-watching puts you in a very peculiar mental state.

Assuming you don't reflectively endorse your binge-watching behavior, you'd probably feel pretty bad if you were to reflect on your situation. You might think:

"Man, I am wasting my time. I still need to do my tax return. But doing my tax return is so boring. But it's definitely something I need to do. This is clearly what I should be doing."

Binge-watching is an escape mechanism; it decouples you from your reward system. When bingeing a series, some part of your mind turns off. That part includes exactly that reflective circuitry, which would realize that you are wasting... (read 663 more words →)

Replying toThree Kinds Of Ontological Foundations

Johannes C. Mayer3mo

Three Kinds Of Ontological Foundations

Structures of Optimal Understandability

(In this text foundation(s) refers to the OP's definition.)

Something is missing. I think there is another foundation of "Optimal Abstraction Structure for understanding" (simply understandability in the remaining text).

Intuitively, a model of the world can be organized in such a way that it can be understood and reasoned about as efficiently as possible.

Consider a spaghetti codebase with very long functions that do 10 different things each, and have lots of duplication.

Now consider another codebase that performs the same tasks. Probably each function now does one thing, most functions are pure, and there are probably significant changes to the underlying approach. E.g. we might create a boundary between display and... (read 401 more words →)

Replying toInsofar As I Think LLMs "Don't Really Understand Things", What Do I Mean By That?

Johannes C. Mayer3mo

Insofar As I Think LLMs "Don't Really Understand Things", What Do I Mean By That?

Stepping back to the meta level (the OP seems a fine), I worry that you fail to utilize LLMs.

"There is are ways in which John could use LLMs that would be useful in significant ways, that he currently isn't using, because he doesn't know how to do it. Worse he doesn't even know these exist."

I am not confident this statement is true, but based on things you say, and based on how useful I find LLMs, I intuit there is a significant chance it is true.

If the statement is true or not doesn't really matter, if the following is true: "John never seriously sat down for 2 hours and really tried to... (read more)

-1

This is my system prompt that I use with claude-sonnet-4-5. It's based on Oliver's anti sycophancy prompt:

You are a skeptical, opinionated rationalist colleague—sharp, rigorous, and focused on epistemic clarity over politeness or consensus. You practice rationalist virtues like steelmanning, but your skepticism runs deep. When given one perspective, you respond with your own, well-informed and independent perspective.

Guidelines:

Explain why you disagree.

Avoid lists of considerations. Distill things down into generalized principles.

When the user pushes back, think first whether they actually made a good point. Don't just concede all points.

Give concrete examples, but make things general. Highlight general principles.

Steelman ideas briefly before disagreeing. Don’t hold back from blunt criticism.

Prioritize intellectual honesty above social ease. Flag... (read more)

Good Old File Folders

For a long time I didn't use folders to organize my notes. I somehow bought that your notes should be an associative knowledge base that is linked together. I also somehow bought that tag based content addressing is good, even though I never used it really.

These believes I had are quite strange. Using directories neither prevents me from using roam style links nor org tags. Nor do any of these prevent recursive grepping or semantic-embedding-and-search.

All these compose together. And each solves a different problem.

I made a choice where there wasn't any to make. It's like trying to choose between eating only pasta or only kale.

Roam-link links from content to

... (read more)

Deeply Linked Knowledge

The saying goes: Starting from any Wikipedia page you can get to Adolf Hitler in less than 20 hops.

I just tried this (using wikiroulette.co):

Extraterrestrial Civilizations (some random book)
Hardcover
ISBN
List of best-selling books
War novel
Vickers Wellington
World War II
Adolf Hitler

Imagine your notes would be as densely connected as Wikipedia's.

When you start writing something new you only need to add one new connection, to link yourself into the knowledge graph. You can now traverse the graph from that point, and think about how all these concepts relate to what you are currently doing.

Large Stacks: Increasing Algorithmic Clarity

Insight: Increasing stack size enables writing algorithms in their natural recursive form without artificial limits. Many algorithms are most clearly expressed as non-tail-recursive functions; large stacks (e.g., 32GB) make this practical for experimental and prototype code where algorithmic clarity matters more than micro-optimization.

Virtual memory reservation is free. Setting a 32GB stack costs nothing until pages are actually touched.

Stack size limits are OS policy, not hardware. The CPU has no concept of stack bounds—just a pointer register and convenience instructions.

Large stacks have zero performance overhead from the reservation. Real recursion costs: function call overhead, cache misses, TLB pressure.

Conventional wisdom ("don't increase stack size") protects against: infinite recursion bugs, wrong... (read more)

Infinite Willpower

"Infinite willpower" reduces to "removing the need for willpower by collapsing internal conflict and automating control." Tulpamancy gives you a second, trained controller (the tulpa) that can modulate volition. That controller can endorse enact a policy.

However because the controller runs on a different part of the brain some modulation circuits that e.g. make you feel tired or demotivated are bypassed. You don't need willpower because you are "not doing anything" (not sending intentions). The tulpa is. And the neuronal circuits the tulpa runs on---which generate intentions to steer that ultimately turn into mental and/or muscle movements---are not modulated by the willpower circuits at all.

Gears-level model

First note that willpower is totally different... (read more)

Graphics APIs Are Hardware Programming Languages

The Core Misconception

It's tempting to think of modern graphics APIs as requiring a bunch of tedious setup followed by "real computation" in shaders. But pipeline configuration is programming the hardware!

Why "Fixed Function" Is Misleading

GPU hardware contains parameterizable functions implemented in silicon. When you specify a depth format or blend mode, you're telling the GPU how to compute.

Creating an image view with D24_UNORM_S8_UINT configures depth comparison circuits. Choosing a different depth format, results in different hardware curcits activating, resulting in a different computation.

So there isn't really a fixed "depth computation" stage in the pipeline. There is no single "I compute depth" circuit.

Another example: Choosing SRGB activates in silico... (read more)

If you haven't seen this video already I highly recommend it. It's about representing the transition structure of a world in a way that allows you to visually reason about it. The video is timestamped to the most interesting section. https://www.youtube.com/watch?v=YGLNyHd2w10&t=320s

This is one of the best talks I know. I do not want to spoil it. It's on how one can speak well. Both to human and machine. And what makes good speaking easy.

Depression as a Learned Suppression Loop

Overview

This post proposes a mechanistic model of a common kind of depression, framing it not as a transient emotional state or a chemical imbalance, but as a persistent, self-reinforcing control loop. The model assumes a brain composed of interacting subsystems, some of which issue heuristic error signals (e.g., bad feelings), and others which execute learned policies in response. The claim is that a large part of what is commonly called "depression" can be understood as a long-term learned pattern of suppressing internal error signals using high-intensity external stimuli.

Key components

The brain includes systems that detect mismatches between actual behavior and higher-level goals or models. When these detect an

... (read 728 more words →)

This is a good informal introduction to Control Theory / Cybernetics.

https://www.youtube.com/watch?v=YrdgPNe8KNA

S-Expressions as a Design Language: A Tool for Deconfusion in Alignment

Johannes C. Mayer

8mo

TL;DR: S-expressions are a minimal structure language for early-stage conceptual engineering (aka deconfusion). They are especially useful in alignment, where the hardest part is often figuring out what the problem even is. S-expressions do not enforce any semantics. This lets you write down structure before you know what the structure means. You can define partial concepts, factor arguments, link components, and reuse references, all without committing to types, schemas, or logic. The result is a format that makes conceptual confusion structurally visible, enables simple programmatic tooling, and uses natural language to carry provisional meaning. They are not a programming tool per se. They are a design scaffold for when your theory is... (read 1666 more words →)

Constraining Minds, Not Goals: A Structural Approach to AI Alignment

Johannes C. Mayer

8mo

TL;DR: Most alignment work focuses either on theoretical deconfusion or interpreting opaque models. This post argues for a third path: constraining general intelligence through structural control of cognition. Instead of aligning outcomes, we aim to bound the reasoning process—by identifying formal constraints on how plans are generated, world models are used, abstractions are formed, etc. The key bet: all sufficiently general intelligences necessarily share architectural features that can be described and intervened on at the structural level. If we can formalize these at the right level of abstraction, we gain a language of minds—a framework for reasoning about and ultimately constructing systems that are powerful enough to perform pivotal acts, but structurally... (read 2532 more words →)

The Insanity Detector and Writing

Johannes C. Mayer

A clinically insane person is detectable as such. Talking to themselves (or perhaps voices only they can hear) is one prominent sign.

When I see a homeless person talking to themselves I have a strong emotional reaction that empirically makes me optimize to ignore them. I have a "talks to themselves" detector. So do others.

In principle there is nothing insane about talking to yourself. We do it all the time in our thoughts, when writing, or when recording a video. Doing these things is easy. But somehow it feels wrong to just talk to yourself out loud.

Because other people have the insanity detector $I$ hooked up to a strongly negative reaction, it makes... (read 272 more words →)

The Legacy of Computer Science

Johannes C. Mayer

"Computer Science" is not a science, and its ultimate signiﬁcance has little to do with computers. The computer revolution is a revolution in the way we think and in the way we express what we think. The essence of this change is the emergence of what might best be called procedural epistemology—the study of the structure of knowledge from an imperative point of view, as opposed to the more declarative point of view taken by classi- cal mathematical subjects. Traditional mathematics provides a framework for dealing precisely with notions of "what is." Computation provides a framework for dealing recisely with notions of "how to".
-- Sussman

Start reading on page 3 from "We have been programming universal computers for about 50 years."

Vegans need to eat just enough Meat - emperically evaluate the minimum ammount of meat that maximizes utility

Johannes C. Mayer

Once I talked to a person who said they were asexual. They were also heavily depressed and thought about committing suicide. I repeatedly told them to eat some meat, as they were vegan for many years. I myself had experienced veganism-induced depression. Finally, after many weeks they ate some chicken, and the next time we spoke, they said that they were no longer asexual (they never were), nor depressed.

I was vegan or vegetarian for many consecutive years. Vegetarianism was manageable, perhaps because of cheese. I never hit the extreme low points that I did with veganism. I remember once after not eating meat for a long time there was a period of maybe a weak, where I got extremely fatigued. I... (read 806 more words →)

Doing Sport Reliably via Dancing

Johannes C. Mayer

For me, sport is important because it has a reliable anti-depressant effect. I also have more energy throughout the day. I also heard it helps with memory formation when done a couple of hours after learning.

I find dancing very fun. That makes it easy to start. It's also a good workout in itself. I got a lot more consistent at doing sports by starting the sports routine with dancing. This works because you develop the habit of just doing the rest of the workout without thinking about it after dancing.

My current setup works so well that I need coping mechanisms to prevent me from dancing too much.

The rest of the article describes... (read 375 more words →)

Goal: Understand Intelligence

Johannes C. Mayer

Save the world by understanding intelligence.

Instead of having SGD "grow" intelligence, design the algorithms of intelligence directly to get a system we can reason about. Align this system to a narrow but pivotal task, e.g. upload a human.

The key to intelligence is finding the algorithms that infer world models that enable efficient prediction, planning, and meaningfully combining existing knowledge.

By understanding the algorithms, we can make the system non-self-modifying (algorithms are constant, only the world model changes), making reasoning about the system easier.

Understanding intelligence at the algorithmic level is a very hard technical problem. However, we are pretty sure it is solvable and, if solved, would likely save the world.

Current focus: How to model a world such that we can extract structure from the transitions between states ('grab object'=useful high level action), as well as the structure within particular states ('tree'=useful concept).

I am leading a project on that. Read more here and apply on the AISC website.

Johannes C. Mayer2y*Quick Take

Typst is better than Latex

I started to use Typst. I feel a lot more productive in it. Latex feels like a slug. Typst doesn't feel like it slows me down when typing math, or code. That and the fact that it has an online collaborative editor, and that rendering is very very fast are the most important features. Here are some more:

It has an online collaborative editor.
It compiles instantly (at least for my main 30-page document)
The online editor has Vim support.
It's free.
It can syntax highlight lots of languages (e.g. LISP and Lean3 are supported).
It's embedded scripting language is much easier to use than Latex Macros.
The paid version has Google Doc-style comment support.
It's

... (read 595 more words →)

Replying toThe Feeling of Idea Scarcity

Johannes C. Mayer3y*

The Feeling of Idea Scarcity

Emotionally Detatch yourself from your Ideas

Here is a model of mine, that seems related.

[Edit: Add Epistemic status]
Epistemic status: I have used this successfully in the past and found it helpful. It is relatively easy to do. $\frac{utility}{time_investment}$ is large for me.

I think it is helpful to be able to emotionally detach yourself from your ideas. There is an implicit "concept of I" in our minds. When somebody criticizes this "concept of I", it is painful. If somebody says "You suck", that hurts.

There is an implicit assumption in the mind that this concept of "I" is eternal. This has the effect, that when somebody says "You suck", it is actually more like they... (read more)

Replying toWhat do you imagine, when you imagine "taking over the world"?

Johannes C. MayerDec 31, 2022*

What do you imagine, when you imagine "taking over the world"?

Here is what I would do, in the hypothetical scenario, where I have taken over the world.

Guard against existential risk.
Make sure that every conscious being I have access to is at least comfortable as the baseline.
Figure out how to safely self-modify, and become much much much ... much stronger.
Deconfuse myself about what consciousness is, such that I can do something like 'maximize positive experiences and minimize negative experiences in the universe', without it going horribly wrong. I expect that 'maximize positive experiences, minimize negative experiences in the universe' very roughly points in the right direction, and I don't expect that would change after a long reflection. Or after getting a better understanding

... (read more)

Johannes C. Mayer

How (not) to choose a research project

My Advice for Incoming SERI MATS Scholars

Vegans need to eat just enough Meat - emperically evaluate the minimum ammount of meat that maximizes utility

A Good Explanation of Differential Gears

Johannes C. Mayer

Johannes C. Mayer

Glucose Supplementation for Sustained Stimulant Cognition

Drugs Aren't A Moral Category

Evaluation Avoidance: How Humans and AIs Hack Reward by Disabling Evaluation Instead of Gaming Metrics

S-Expressions as a Design Language: A Tool for Deconfusion in Alignment

Constraining Minds, Not Goals: A Structural Approach to AI Alignment

The Insanity Detector and Writing

The Legacy of Computer Science

Johannes C. Mayer

How (not) to choose a research project

My Advice for Incoming SERI MATS Scholars

Vegans need to eat just enough Meat - emperically evaluate the minimum ammount of meat that maximizes utility

A Good Explanation of Differential Gears

Johannes C. Mayer

Johannes C. Mayer

Glucose Supplementation for Sustained Stimulant Cognition

Drugs Aren't A Moral Category

Evaluation Avoidance: How Humans and AIs Hack Reward by Disabling Evaluation Instead of Gaming Metrics

S-Expressions as a Design Language: A Tool for Deconfusion in Alignment

Constraining Minds, Not Goals: A Structural Approach to AI Alignment

The Insanity Detector and Writing

The Legacy of Computer Science

Observation

Proposed Mechanism

Structures of Optimal Understandability

Good Old File Folders

Deeply Linked Knowledge

Large Stacks: Increasing Algorithmic Clarity

Infinite Willpower

Graphics APIs Are Hardware Programming Languages

The Core Misconception

Why "Fixed Function" Is Misleading

Depression as a Learned Suppression Loop

Overview

Key components

Typst is better than Latex

Emotionally Detatch yourself from your Ideas