6 ms·

Human Memory

78 points by andyxor 5y ago

etherio 5y ago
Interesting! I learned a lot of new things about memory when I did the "learning how to learn course" [0]. I definitely recommend it for people trying to learn vast amounts and use their memory effectively. The link's more specific honing on memory itself seems more in depth however. [0]: https://www.coursera.org/learn/learning-how-to-learn https://www.coursera.org/learn/learning-how-to-learn
andyxor 5y ago
If you're interested in this you may like the instructor's recent paper "Episodic memory: mental time travel or a quantum `memory wave' function?" https://psyarxiv.com/6zjwb https://psyarxiv.com/6zjwb as a side note I couldn't find a reference to hippocampal grid cells (and their role in memory) in the course materials, but may be I missed it https://www.sciencedirect.com/science/article/pii/S2352154616302819 https://www.sciencedirect.com/science/article/pii/S235215461... also the associative memory section missing many important models, like sparse distributed memory, may be worth a PR https://en.wikipedia.org/wiki/Sparse_distributed_memory https://en.wikipedia.org/wiki/Sparse_distributed_memory
askthereception 5y ago
The first question on the first problem sheet reads: > Describe one aspect of memory that you are absolutely certain is true. Explain how you know– be as specific as possible. Any thoughts?
- TrainedMonkey 5y ago
  Memory is associative. I know this because someone told me a long time ago and then I've seen many studies which has shown that. I am sure some confirmation bias was at play. I specifically recall two types of studies. First one is that giving a hint of any specific part of memory aids in recalling of a whole memory. People have used this to construct memorization techniques, such as memory palace. Second one is that memory created during an emotional state is easier to remember when emotional state is recreated. Additionally I have not seen any studies which disprove memory associativity claim, although I am sure there are some and I just have not looked hard enough.
- amalcon 5y ago
  Memory is fallible. I know because I remember my memory failing me. If that meta-memory is accurate, then it is evidence that my memory is fallible. If it is inaccurate, then I remember something that didn't happen, so my memory is fallible anyway.
  - amboo7 5y ago
    Are you sure you remember when your memory is faillible? ;-) (Mine is perfect, I don't remember I ever forgot something.)
- hypertele-Xii 5y ago
  Memory is subconscious. You may recall things without consciously thinking about them, and you may fail to recall things when specifically thinking about them. Conscious thinking is good at logically synthesizing information as it's needed, but some property of it forbids it from actually storing any information. Real world data just doesn't organize neatly; The subconscious is fine being a mess. Try this experiment: Ask your subconscious to recall a specific memory you know you have. Then, stop thinking about it entirely. Within an unspecified length of time (could be minutes, hours, days), you'll spontaneously recall details about that memory you didn't know you had.
emmanueloga_ 5y ago
Thanks for sharing, haven't seen this before, looks like a great resource. Two other things I think about when I think about Human Memory: * "Funes the Memorious" [1], a short story from J.L.Borges that describes what it may look like to truly have perfect memory. * The writings of Piotr Wozniak [2], author of Supermemo. -- 1: https://en.wikipedia.org/wiki/Funes_the_Memorious https://en.wikipedia.org/wiki/Funes_the_Memorious 2: https://supermemo.guru/wiki/Piotr_Wozniak https://supermemo.guru/wiki/Piotr_Wozniak
Dig1t 5y ago
Well this is amazingly cool, are the lectures for this course available anywhere? I would love to have lectures like this available on something like MIT OpenCourseWare
- andyxor 5y ago
  there are all on youtube, just search for "recording" links on that page
  - Dig1t 5y ago
    Oh I missed that, thank you!
aliasEli 5y ago
I have not really looked into its contents, but a course that suggests Christopher Nolan's "Memento" as additional material certainly sounds interesting.
- closed 5y ago
  Weirdly enough, Momento is a big hit with a good chunk of memory researchers, and had been used as a stimulus in memory experiments! https://www.aalto.fi/en/news/film-memento-helped-uncover-how-the-brain-remembers-and-interprets-events-from-clues https://www.aalto.fi/en/news/film-memento-helped-uncover-how...

78 points by andyxor 5y ago

Interesting! I learned a lot of new things about memory when I did the "learning how to learn course" [0]. I definitely recommend it for people trying to learn vast amounts and use their memory effectively. The link's more specific honing on memory itself seems more in depth however. [0]: https://www.coursera.org/learn/learning-how-to-learn https://www.coursera.org/learn/learning-how-to-learn

andyxor 5y ago

If you're interested in this you may like the instructor's recent paper "Episodic memory: mental time travel or a quantum `memory wave' function?" https://psyarxiv.com/6zjwb https://psyarxiv.com/6zjwb as a side note I couldn't find a reference to hippocampal grid cells (and their role in memory) in the course materials, but may be I missed it https://www.sciencedirect.com/science/article/pii/S2352154616302819 https://www.sciencedirect.com/science/article/pii/S235215461... also the associative memory section missing many important models, like sparse distributed memory, may be worth a PR https://en.wikipedia.org/wiki/Sparse_distributed_memory https://en.wikipedia.org/wiki/Sparse_distributed_memory

askthereception 5y ago

The first question on the first problem sheet reads: > Describe one aspect of memory that you are absolutely certain is true. Explain how you know– be as specific as possible. Any thoughts?

TrainedMonkey 5y ago
Memory is associative. I know this because someone told me a long time ago and then I've seen many studies which has shown that. I am sure some confirmation bias was at play. I specifically recall two types of studies. First one is that giving a hint of any specific part of memory aids in recalling of a whole memory. People have used this to construct memorization techniques, such as memory palace. Second one is that memory created during an emotional state is easier to remember when emotional state is recreated. Additionally I have not seen any studies which disprove memory associativity claim, although I am sure there are some and I just have not looked hard enough.
amalcon 5y ago
Memory is fallible. I know because I remember my memory failing me. If that meta-memory is accurate, then it is evidence that my memory is fallible. If it is inaccurate, then I remember something that didn't happen, so my memory is fallible anyway.
- amboo7 5y ago
  Are you sure you remember when your memory is faillible? ;-) (Mine is perfect, I don't remember I ever forgot something.)
hypertele-Xii 5y ago
Memory is subconscious. You may recall things without consciously thinking about them, and you may fail to recall things when specifically thinking about them. Conscious thinking is good at logically synthesizing information as it's needed, but some property of it forbids it from actually storing any information. Real world data just doesn't organize neatly; The subconscious is fine being a mess. Try this experiment: Ask your subconscious to recall a specific memory you know you have. Then, stop thinking about it entirely. Within an unspecified length of time (could be minutes, hours, days), you'll spontaneously recall details about that memory you didn't know you had.

emmanueloga_ 5y ago

Thanks for sharing, haven't seen this before, looks like a great resource. Two other things I think about when I think about Human Memory: * "Funes the Memorious" [1], a short story from J.L.Borges that describes what it may look like to truly have perfect memory. * The writings of Piotr Wozniak [2], author of Supermemo. -- 1: https://en.wikipedia.org/wiki/Funes_the_Memorious https://en.wikipedia.org/wiki/Funes_the_Memorious 2: https://supermemo.guru/wiki/Piotr_Wozniak https://supermemo.guru/wiki/Piotr_Wozniak

Dig1t 5y ago

Well this is amazingly cool, are the lectures for this course available anywhere? I would love to have lectures like this available on something like MIT OpenCourseWare

andyxor 5y ago
there are all on youtube, just search for "recording" links on that page
- Dig1t 5y ago
  Oh I missed that, thank you!

aliasEli 5y ago

I have not really looked into its contents, but a course that suggests Christopher Nolan's "Memento" as additional material certainly sounds interesting.

closed 5y ago
Weirdly enough, Momento is a big hit with a good chunk of memory researchers, and had been used as a stimulus in memory experiments! https://www.aalto.fi/en/news/film-memento-helped-uncover-how-the-brain-remembers-and-interprets-events-from-clues https://www.aalto.fi/en/news/film-memento-helped-uncover-how...