What surprises me (and makes me most skeptical) is that they can remember days by date. Did these people actually know what the date was every year of their lives? This implies to me that it is not just a camcorder-like memory that these people possess, but an efficient, numerical search ability on those memories. It makes me wonder how much one could duplicate this ability by purposefully associating numbers with days in a hierarchical structure (analogous to building an explicit tree structure on them).
Also fascinating to me is how this links with OCD-like behavior. My memory has always been considered very good (though not like these people), and I had extremely mild OCD during my preteen years, which changed its form when I became a teenager until I stopped manifesting most symptoms.
I also wonder what would happen if these people were asked to gather days by an arbitrary filter. For example, on which days during 1999 did it rain? Would they have to step through day-by-day and check each day, or would they be able to instantly run through the rainy days? The latter seems almost too powerful.
I agree.. Why did they keep saying "it was a Monday" or whatever? Seemed pretty random/uncalled for. Makes me think of they way things are organized in their brains.
I also wonder how they're not very high-performing and successful individuals. It seems to me an "endless" memory could be hugely beneficial, beyond the many (unethical) scheme they can exploit.
These people's "perfect" memories are purely autobiographical. They cannot memorize large lists of numbers quickly, or recite the entirety of The Bible from memory, or excel at any other task that requires rapid, verifiable memorization. Nothing about this is inconsistent with these people simply spending, for whatever reason, a considerable amount of effort to memorize and catalog the minutiae of their lives, but otherwise having normal mental faculties.
What is the decision procedure for if some memory is "autobiographical"? I don't understand why, if the person read a long list of numbers, they wouldn't be able to remember them, yet they can remember a long list of things they ate, the weather, or what TV shows they watched.
I wish the article went into more detail about how "perfect" these memories are. Of course, other than big events (series premieres, awards ceremonies) and well-recorded things like the weather, it's very difficult to verify their accuracy.
I have other questions. Is there an "alertness level" required for memories to be committed? I've gone several days in a row just sitting around my house lazy and completely unaware of what day of the week or date was. If I had this "perfect memory," would I be able to recall decades later the day of the week or the date?
In at least one instance, it turned out the woman was able to remember everything she did because she spent a lot of time thinking about and reviewing what she did. She didn't spend nearly as much time memorizing numbers and wasn't as good at it.
I agree with the sentiment of your point, but this does represent "normal mental faculties" in any way. They are clearly extraordinary in this one sense and that alone doesn't allow them to be classified as normal.
I watched this spot on 60 minutes and I am not convinced that these people are normal with respect to mental health. They had a very odd affect at times. Still, it was amazing to watch people recall things that happened over 20 years ago with clarity and precision.
When I was in high school, I could recite nearly 1,500 digits of pi from memory. This wasn't because I had number-memorizing superpowers, it was because I was a geek and I spent time memorizing numbers. If you look at interviews with top competitive mnemonists, you find very few who claim to have (let alone actually have) any sort of "photographic memory" or even autistic savantisms. It's just a mundane subject with strategies and tricks. People don't close their eyes and see numbers, they break them up into convenient chunks, look for symmetries and patterns, and memorize them.
You can't underestimate what people can accomplish when they have severe abnormalities in their interests. Calendar-counting to compute days of the week is not hard. Weather patterns are easily mentally compressible, with natural chunking points from the seasons. Most people's work schedules are highly regular, so it's easy to break weeks into patterns and exceptions. And so on. Everything that's demonstrated in this article is well within the realm of feasibility for a sufficiently dedicated person without superpowers.
I think the point he was trying to make is this: would you consider someone who could completely crush you at, say, basketball, to have "abnormal powers"?
I'm not talking pros, I'm talking about people who've played sports all their lives. They can probably beat most people at their chosen sports. And a lot of perfectly "normal/average" people are really good at one sport or another. We don't talk about them as if they're "abnormal".
Haven't done it, but I'm guessing that memorizing 1500 digits of pi is easier and less time consuming than getting to a decent level at some sport. Except that for most people, playing sport is fun, so they don't think about it as "extreme concentration". For the op, memorizing digits of pi was fun (I'm guessing?).
I think you're completely misunderstanding the thrust of the article. This ability goes far beyond memorizing knowledge in one specialized area. These people go through hours (up to 8 at a time) of testing about the everyday mundane events that have happened as well as the extraordinary things - there is no way for them to trick their testers. In the 60 minute piece, they could recall amazing detail about days from 20 years ago. This goes far beyond memorizing weather patterns and reconstructing information according to the regularity of a work schedule, etc.
I think what a lot of people are noticing here is that this extraordinary ability has not been commensurate with extraordinary accomplishments and it leads them to downplay the significance...
I'm not saying their abilities aren't "extraordinary" or at least very impressive compared to what one might naively imagine. I am more saying that people who spend endorsement efforts in other areas also tend to gain enormous abilities and so these are "normal" extraordinary abilities.
The point others have raised about these people not having abilities in other areas goes beyond this "unique talent" "not transferring". The point is that the people with these "abilities" spend a fair portion of their waking hours maintaining them. The memories may be accurate but they're "honed", time is spent codifying them.
Most specifically, the memories are clearly very different anything "photographic".
Just consider, if a person actually a video tape of their entire life, using said tape to answer questions about "what happen on day X" would take a few hours of mental processing. These people answer such questions instantly.
They've made themselves "experts" on one subject, themselves, and the ability give instant answers on that subject is a clue to this...
'Just consider, if a person actually a video tape of their entire life, using said tape to answer questions about "what happen on day X" would take a few hours of mental processing.'
You're making a huge pile of unsupported assertions, but this is the most egregious. You have the outline of an interesting theory, but little more. For one thing I see no evidence that there is any conscious effort made to maintain these memories, in many cases the people express every desire not to have this happen, yet it happens anyhow. At the point where your "making themselves experts" happens at a purely automatic level and against their conscious will you've stretched the words beyond any reasonable meaning, and I'm not sure your theory can even in principle be converted into something that corresponds to the real world.
The article states that some of these people don't even want to remember everything - so there obviously are people who don't spend a lot of time exercising this skill but who simply have it.
But how many six-year-olds would succeed at memorizing every event in their their life if they ever decided to do so? Since there wasn't any environmental stimulus for them to do so I suspect a genetic reason for their ability.
They also never mentioned that they somehow decided to spend a lot of time cataloging their memories. It seems to be totally effortless for them.
I'm unconvinced - what verifiable memory feats are presented? If asked "what were you doing on such and such a date 20 years ago", they can say anything and who can contradict them?
Perhaps they simply are good storeytellers. Which is a feat in itself.
This article skips a very important aspect of these subjects: can they choose a topic and learn it inside and out and not forget it? For example, if they were interested in, say, botany, could they use their fantastic memory to remember detailed characteristics of every plant species they study?
All they seem to remember is what the did on a given day, or who won a sporting event or the weather.
I've read about people having this ability in the past, but it was usually related to either brain damage or idiot savancy. I think scientists have just starting studying "normal" people with the ability.
I'm pretty sure Sherashevsky (studied in the early 1900s) falls into the 'normal' people category. He certainly had problems but would be a very high functioning idiot savant at absolute worst.
In short: it turns out there are (few) people that have almost perfect memory, even 20 years back. And that this type of memory is totally new to science.
I read in my childhood about man who remember everything. He worked as a reporter and never wrote down any tasks editor gave him. That man told the editor that he won't forget and he did not forget anything, really. So editor send that man to psychologist who discover that he had perfect memory.
)the story was about 60-s or so)
The perfect rememberer had a mild case of synestesia, but he used it to his own advantage. He put so many hooks on anything that he couldn't forget even if he try. He remembers conversation by their content, sound, color, texture and taste (and much more). So every word has its own color, texture and taste, they make a web of them, much like labyrinth or city map.
When I need something I shouldn't forget, I also try to put as many markers as I could. It hepls sometimes. ;)
Radiolab (a podcast that any HN reader would love) did a piece once about a Russian who had perfect recall of his life and could also memorize endless lists of numbers and Dante's Inferno in Italian (and didn't speak Italian). Unfortunately, he eventually went mad because his brain couldn't make sense of all the interconnections.
Emphasis on almost. As would be expected, their memory isn't actually perfect.
Thermodynamics/information theory put a limit on how much information you can pack into a certain area/mass, and with our knowledge of biology we already know that the vast majority of the grey matter in our skulls isn't used for information storage.
In other words, this is interesting from a medical perspective, but not a computer science/mathematics/physics perspective.
On the other hand, if we could get ahold of a human with a perfect memory, we could easily establish (at least on the basis of orders of magnitude) how much concious-level information can be stored in the brain. This could grant tremendous insight into the functioning of our brains, as it would also give us a good idea of the "size" (in terms of neural connections, for example) of memories, thoughts and so forth as they are stored.
We already have information-theoretic bounds on this sort of thing based on a physical understanding of the number of neurons and synapses in a person's skull, but they're very loose. It's like trying to estimate the storage available in a chip from the number of transistors on the die. What kind of RAM does it contain? How much is support circuitry? Does it contain RAM at all, or is it something more like a dataflow CPU? If we can fit a high-entropy sequence of N distinct digits into a brain composed of M neurons we might presume that storing each digit involves M/N neurons. If N is close to or greater than M, we might instead guess that data must be stored in-flight, which leads us down different avenues of investigation.
But this is all really a thought experiment, anyway...
Remembering the feeling of a drug, emotion or experience isn't the same as experiencing it. Also (trust me on this) regardless of how good your memory is, once you start drinking, smoking pot, taking acid, etc. it's not going to be quite the same for some time, if ever, as your habits of recall and storage will change.
Trivia TV shows, no doubt. But I bet even these people's memories have a limit.
Also fascinating to me is how this links with OCD-like behavior. My memory has always been considered very good (though not like these people), and I had extremely mild OCD during my preteen years, which changed its form when I became a teenager until I stopped manifesting most symptoms.
I also wonder what would happen if these people were asked to gather days by an arbitrary filter. For example, on which days during 1999 did it rain? Would they have to step through day-by-day and check each day, or would they be able to instantly run through the rainy days? The latter seems almost too powerful.