Hacker News Reader

Comment by firasd | Hacker News Reader

Doesn’t the concept of a subconscious, or inner voice, invalidate any tight analogies to Turing Machines? What logical process is putting the Fur Elise into my head (earworms etc)

I think this article is right that definitions really matter here cause one could argue “CTM is trivial, because we can describe almost any physical system as executing computations.” So we have a dual problem here—not just the difficult discussion of ‘what’s going on in minds’, but also ‘what is computation’.

View on HN

Wonder if the subconscious could simply be another LLM with a higher temperature setting…

Have long thought that creativity is little more than a finely tuned filter driven by noise…

Agreed. Well said.

Trying to maintain that our mind is unreachable by the laws of logic is absurd.

If reachable by laws of logic, then it can be stated. If it can be stated, it can be executed.

Its funny that the very act of chaining sentences is what logic is. And every human thought can be written as a sentence.

> every human thought can be written as a sentence.

No they can't. I have things in my brain that I don't have words for, that I'm able to find things other people have done to help me express them. You can attach sentences to music and art, but a picture is worth a thousand words. The thoughts on my head when I saw the latest Banksy, for example. Maybe I'm abberationally ineloquent, but I don't think I'm that exceptional.

Yes they can. Retreating to "I feel things I can't say" arguments that cannot be falsified is a cope.

Either that or you didn't put effort into writing it down or you lack the capability to do it.

You can even write sentences like "cows have wings and smells green". Regular thoughts are feeble by comparison in terms of complexity.

Claiming "a picture is worth a thousand words" may work out in family gatherings but irrelevant and meaningless in a computation discussion. Picture is represented by concrete symbols.

Having an internal representation doesn’t imply an English sentence exists. Temple Grandin's written about "thinking in pictures". Language is lossy. For a doctor, the question, how much does it hurt, on a scale from 1-10? is the best we can do.

Appealing to some authority won't rescue you as information theory doesn't work by arbitrary authority figures.

Thinking computational theory is about hurt is a joke. Resorting to feelings signifies lack of technical substance.

Language is precise or lossy depending on granularity.

Haven't you seen a png file in a computer. Visuals are symbols buddy. Hence representable in language.

> Visuals are symbols buddy.

that's a weaker claim than "every human thought can be written as a sentence."

I would agree, but I recently read https://transformer-circuits.pub/2026/workspace/index.html which is quite interesting in it's own right.

What are your thoughts on this?

Are LLMs even covered by ‘computational theory of mind’? Cause if I say “I like to bake” the LLM is only able to recognize and respond to that sentence due to the neural network making emergent connections between the tokens in the words—very different from the kind of computer programs these guys seem to have been thinking about.

I think many cognitive scientists think language is a core part of human cognition, and that LLMs were a major advance in AI/cognitive science due to connecting AI algorithms with language production, but LLMs aren’t considered (or even aiming) to be accurate models of human cognition

Infuriatingly, computation is subjective. The experience of the inputs, the results, and the processes depends on qualia. So does logic itself.

Automating the processes doesn't change that. There's a very large number of possible processes and operations, but only a small number are perceived as valuable.

Reply on news.ycombinator.com