Native fluency: only 1.5 MB of information needed?

[PeterMollenburg]

The brain is not computer, nor does it use bits. It’s not surprising that the article referenced the 1958 von Neumann “Computer and The Brain” for its estimation.

All these discussions about how many bits a word takes ... we process language as sounds, not as text.

I agree, it’s as abstract as judging wine quality based on how many bananas a monkey eats. A good use of one’s time and energy

[DaraghM]

Brilliant. So all of my language ability fits into one single track on a language course, and here I am messing around with gigabytes of CD's and MP3s.

[tommus]

we process language as sounds, not as text.

There are lots of people who learn a second language only to read it (and perhaps to write it) but who rarely if ever hear or speak it. So sound is not really involved in storing the information. Yes, we process the sounds, but the knowledge and memory of language is not stored as a sound. There aren't thousands of little sounds buzzing around in our memories. The necessary knowledge and memory must be stored in the brain as some sort of code, and storage of code requires fairly accurately known numbers of "bits" of data. So I think the amount of "memory space" that it takes to store whatever we learn as language (even if we learn it by sound) is interesting. Besides the data for the vocabulary, we have to store knowledge of what things look like (an elephant, a tree, etc.) but that is not counted in the amount of memory required for fluency. A person would have all that recognition of things, etc. even if they grew up totally isolated and never learned a language. Agreed; there would be a difference in the amount of bits require to store the sound of a word compared to the amount required for only the spelling of a word. But there would be a correlation (long sounds correlate to long words). So if a person learns to read, write, listen and speak a language, perhaps the storage requirements are something like 4 x 1.5 MB, plus overhead for grammar, expressions, etc.

[DaveAgain]

we process language as sounds, not as text.

There are lots of people who learn a second language only to read it (and perhaps to write it) but who rarely if ever hear or speak it. So sound is not really involved in storing the information. Yes, we process the sounds, but the knowledge and memory of language is not stored as a sound. There aren't thousands of little sounds buzzing around in our memories. The necessary knowledge and memory must be stored in the brain as some sort of code, and storage of code requires fairly accurately known numbers of "bits" of data. So I think the amount of "memory space" that it takes to store whatever we learn as language (even if we learn it by sound) is interesting. Besides the data for the vocabulary, we have to store knowledge of what things look like (an elephant, a tree, etc.) but that is not counted in the amount of memory required for fluency. A person would have all that recognition of things, etc. even if they grew up totally isolated and never learned a language. Agreed; there would be a difference in the amount of bits require to store the sound of a word compared to the amount required for only the spelling of a word. But there would be a correlation (long sounds correlate to long words). So if a person learns to read, write, listen and speak a language, perhaps the storage requirements are something like 4 x 1.5 MB, plus overhead for grammar, expressions, etc.

When we read, letters/groups of letters are mapped to sounds, and then handled by the brain as sound.

https://forum.language-learners.org/vie ... 01&p=84766

[VyingEye]

I think it's worth pointing out that this paper is written from an information theoretic approach.
Meaning the 1.5MB of data is information one extracts after a hard day's work every day over 18 years.
Not at all like reading a 200,000 word book.
More like taking 12.5 million left or right turns while flying through space to end up on planet Earth.
You go through terabytes of data every day over those 18 years to extract those average 1900 bits for the day.
Same way you can download Google Translate for your language pair and it takes only ~100MB, that doesn't mean that's how much data was necessary to train that model, which is easily 10000 times higher. Not to mention the time and work effort to process all of it.

To quote the paper itself:

To put our lower estimate in perspective, each day for 18 years a child must wake up and remember, perfectly and for the rest of their life, an amount of information equivalent to the information in this sequence,
011010000110100101100100011001
000110010101101110011000010110
001101100011011011110111001001
100100011010010110111101101110

And that's the minimum estimate of 120 bits, average estimate is 1900 bits, and upper estimate is 6200 bits (50 sequences like the one above) to be extracted and memorized perfectly and for the rest of your life every day over 6570 days.

P.S. Also I think it's interesting to point out that ~96% of all the work goes into lexical semantics, a.k.a. learning the meaning of things in the concept space.

[tommus]

When we read, letters/groups of letters are mapped to sounds, and then handled by the brain as sound.

and stored in our memory as bits.

[Lawyer&Mom]

The brain is not computer, nor does it use bits. It’s not surprising that the article referenced the 1958 von Neumann “Computer and The Brain” for its estimation.

All these discussions about how many bits a word takes ... we process language as sounds, not as text.

And everyone knows sound files are larger than text files!

[Cainntear]

All these discussions about how many bits a word takes ... we process language as sounds, not as text.

And everyone knows sound files are larger than text files!

Just run it through pkzip, then pkzip it again, and again as nauseum until it fits.

[half entity]

Proof that English is easier.

German
Bundespräsidentenstichwahlwiederholungsverschiebung

Great word. I didn't know it. Reminds me of the Swiss minister who couldn't stop laughing in front of the parliament because of the bureaucratic language in his document.

[Eriol]

All these discussions about how many bits a word takes ... we process language as sounds, not as text.

And everyone knows sound files are larger than text files!

Just run it through pkzip, then pkzip it again, and again as nauseum until it fits.

I'm pretty sure the brain runs a lossy compression algorithm. At least mine.