Long-term memory vs computer memory

Someone shared this article with me:

This prompted some thoughts:

1. It says long-term memory is unlimited. I don't see how this is true when there is a physical limitation on neurons. Estimates put brain storage at about 2.5 million gigabytes. Meanwhile the internet, or the cloud, currently houses about 40 trillion gigabytes, and is constantly increasing. That’s 16 billion times as much as one brain. Or more than twice the brain storage of the entire world population. Perhaps a brain's storage is sufficient to store information for one lifetime of existence. But it is also trims down unused neural pathways, which means...

2. Memory is not permanent. I have learned Japanese using Anki, one of the apps suggested in the article. But I know from experience that does not guarantee I will recall the information quickly when needed. There is also a lot of energy required to get that information stuck into long-term memory, and then keep it there.

3. Memories change. They can get reinterpreted when recalled, and stored again with updated information. For example, when someone is telling a story, they may exaggerate a little. When they recall it next time, they can remember the exaggeration as fact, and then exaggerate a little more. So next time, the story is even more warped. Memories are not reliable, because the brain is not trying to store an accurate representation of facts. It uses filters to save only the most rememberable or pertinent parts. Memories get corrupted by natural human biases. Meanwhile, information that is stored on a computer system and redundantly copied in the cloud is more likely to be preserved accurately. This would be important for sourcing information and keeping the original context.

Short-term working memory is the bottleneck for working with information, for both computer systems and long-term memory. However, second brain apps also allow you to expand this working memory. Think multiple tabs. You can “freeze” what you’re working on by navigating to a different page, then come back to pick up right where you left off, at any time.

Some types of knowledge are a better fit for long-term memory. Such as language learning. It wouldn’t work out well for me to be opening up Chinese class notes in the middle of trying to have a face-to-face conversation with a Chinese person.

For other types of information, long-term memory can be used to store shortcuts or trailheads to the second brain. By using a second brain to parse, synthesize, and distill knowledge, I can save my first brain the effort of remembering all the different parts that make up the final idea.

I don’t always remember what I write. But it surfaces when I seek it out. If I didn’t have a second brain, ideas that I didn’t write down would likely be lost forever. A second brain augments the first brain. Effectively using either one are separate skillsets that must be developed.

Reflecting back on what I used to write the above:

I googled the facts about brain and internet storage. I have probably thought about brain size in computer storage units before, but it was not something I remembered.

The facts about the way memory works I pulled directly from my brain. I had done a lot of research on learning how to learn languages, or learning anything in general. Those ideas just stuck without me trying to force them into long-term memory through space-repetition flashcard reviews. I know I took some notes when learning this, so if I needed to confirm something, I would look up the source. But for the most part, I’m going by what I remember, and may be incorrect.

Oftentimes, open tabs that are not relevant to whatever I’m working on feel like they clutter my brain’s working memory. But I feel safe in storing them in Raindrop bookmarks, knowing they will be retrievable should I ever want to come back to them.