amtrack Wrote:Learning is not overly complicated. If you do it enough, you'll eventually learn it. It is really that simple. How much you need to do it largely depends on how hard a concept it is for you to grasp. But once you "learn it," forgetting it is extremely difficult. All I'm saying is that the SRS algorithm itself does absolutely nothing for you. What helps you learn is that you "just do it". Does anyone honestly think that doing those reviews 3 days later gives them some kind of learning advantage? Of course not. That's just silly.

To be blunt, SRS is basically "made up BS," but it works in that people believe in it enough to stick with their flash card repetitions. Its nothing more than doing flash cards. Thats it. There's no magic in it beyond that. The only difference is that people think these flash cards are "magic" so they do them vigorously. If you did normal flashcards with as much zeal your results would basically be the same.

Artificially spacing things does no more for your memory than encountering the same fact tens of times a day over several days. This is easily found to be true when examining your own memory and thinking of how you have come to learn as much as you do. In this way you come to see that SRS is the usual "scientific hogwash" that is made up to popularize a method.

There is a reason SRS isnt more popular. There are very very intelligent people in charge of learning, and teaching at universities. They recognize that it is basically bologna, which is why its not widely published. Get the right amount of practice and you learn. Don't get the right amount of practice and you don't. There are thousands of ways to get in your practice, so anything works as long as you do it.

I am sorry but this is so wrong it hurts.

Spaced repetition has been shown more effective than cramming in the literature.

The learning system is horribly broken. I remember maybe 15% of classes I took more than a year ago that I no longer use. I couldn't pass many of the AP tests I got 5s on 5 years ago.

Spacing does do more to strengthen memory than cramming, which is why an hour a day over 10 days lets you learn much more than ten hours in one day.

I am no proponent of SRSing everything, but the SPACING EFFECT (not the software, the phenomenom) is backed up strongly in science and I would encourage everyone to ignore any advice you have on the topic.

P.S. If you go to a really good university you will find the professors are chosen primarily for knowing their subject, not for having any clue about how to teach.

amtrack Wrote:"just do it", "made up BS," "magic", "scientific hogwash",
There are very very intelligent people in charge of learning, and teaching at universities. They recognize that it is basically bologna.

If we list out your main points, your argument seems very biased. :O
Most people on this forum would easily disagree with your viewpoint.

In support of your argument, I'll say that it is always best to learn and understand something the first time you encounter it.

Aspiring Wrote:In support of your argument, I'll say that it is always best to learn and understand something the first time you encounter it.

That's also true, but he erroneously dismisses the importance of the spacing effect. There are many aspects of learning. Linking to previous knowledge helps stabilize the memories. Which is why it's easier to memorize a book in a language you know than in one you don't know. And why it's easier to memorize something you understand than something you don't.

I once read a Neuroscientist's Guide for Future Teachers (I can't find it at the moment). There were a lot of good points. How we should link concepts and patterns to ones the students already know, better stabilizing the whole thing. They also mentioned that returning to previously learned material after a break also strengthens it (spacing effect).

Understanding knowledge, finding patterns, using mnemonics, using spacing all help remembering. If one wants to tie their hands behind their back and dismiss the usefulness of any one method, that's their own choice.

If the people in charge of language curricula are so smart and knowledgeable on the topic of learning a language, why don't I know Spanish? I took a course designed by one of these learning geniuses (and it was actually a pretty good class compared to others), but I didn't really learn anything. The only reason any of it stuck was that the pace was so slow that a semesters worth of material isn't even enough to really speak.

Honestly, I think the main point to SRS is that you waste less time reviewing the words you can remember and more time on the words you can't recall. It's not magical, but it does stream line flash card studies and, being a program, doesn't take up a whole shoebox. Trying to manage that many physical cards just isn't something that most people find appealing.

Studying outside of reading is less efficient (I've had cards take a couple of months to really sink in, but I can learn a new word from a book after only a couple of encounters; less in English), but it's a lot easier to study cards when you're just starting out compared to looking up every single word you don't know every time you encounter it.

I guess the point I'm trying to make is, SRS has it's place when you're starting out, but once you can understand enough native material to learn in context you should probably stop using it as much.

@oefirouz
Any memory tool, such as spacing or mnemonics, is used to create & strengthen associations with memories you already know. Spacing is, in a way, association with memories from the past.


Note: I don't actually agree with amtrack. I just wanted to provide one point that would help validate his argument.

&I like how he used an oxymoron to prove himself.
"Scientific hogwash".

Aspiring Wrote:If we list out your main points, your argument seems very biased. :O
Most people [strike] on this forum[/strike] would easily disagree with your viewpoint.

In support of your argument, I'll say that it is always best to learn and understand something the first time you encounter it.

Biased in what fashion? I don't recall showing any bias towards anything. The SRS claims just remind me a lot of things like the "shake weight". Sure some scientists say this works, but it is in no way widely endorsed by the scientific community.

And if I had all the room to type in the world, I'd tell you why I agree with you. I was heading that direction, but it'd take me too much space.

oefirouz Wrote:Spacing does do more to strengthen memory than cramming, which is why an hour a day over 10 days lets you learn much more than ten hours in one day.

I am no proponent of SRSing everything, but the SPACING EFFECT (not the software, the phenomenom) is backed up strongly in science and I would encourage everyone to ignore any advice you have on the topic.

Firstly, nowhere did I mention cramming. I thought it was mostly common sense that you cannot force yourself to learn something, which is the definition of cramming. Two, there is nothing I said that is incorrect. Examine your memory and tell me I am wrong. Do it, I dare you.

I can bet you several things. You know less from "spacing" than you do from constant exposure. Two, someone actively immersed in a language can learn more than they would by "spacing" out a limited exposure. Three, you can change the spacing in SRS at your leisure and it would still work for you. Four, you can review early and achieve the same or better in terms of results. Five, despite spacing, there are some items you just can't seem to remember.

The 5 points above illustrate my main argument: artificially spacing content does not strengthen your memory. It "maintains" your memory..if you space it just right (impossible btw), but it certainly doesn't make it stronger. If we apply a little bit of reason its not hard to come to this conclusion.

You have a variety of things that happened merely once, but you will never forget. Why is that? You have a variety of words you read or heard only once, but again, will never forget. Why is this? This alone proves memory is certainly not a product of some spacing algorithm. Any psychologist could tell you that your strongest memories have absolutely nothing to do with spacing.

EDIT: There are ways to strengthen your memory using SRS, but you have to use it properly.

Mehhh^ biased in an opinionated fashion.

But, you proved your point this time . ^o^

Aspiring Wrote:Mehhh^ biased in an opinionated fashion.

But, you proved your point this time . ^o^

Thank you! I see so basically biased "against" SRS instead of biased "towards" something else lol

Stian Wrote:There is also a concensus on this forum that doing English->Japanese cards is ineffective.

For probably 5 odd years my Anki deck had both E->J and J-E cards added for each. Only recently did I change to having only J-E.

Not doing that from the beginning is probably the biggest mistake in my Japanese study I ever made. Had I only had J->E from the beginning the reviews would have remained much easier leaving me to add far more vocabulary, and I'm sure I'd be far better at Japanese than I am now. Oh well....

At least I'm not making the same mistake for Mandarin =)

NightSky Wrote:Not doing that from the beginning is probably the biggest mistake in my Japanese study I ever made. Had I only had J->E from the beginning the reviews would have remained much easier leaving me to add far more vocabulary, and I'm sure I'd be far better at Japanese than I am now. Oh well....

At least I'm not making the same mistake for Mandarin =)

idk I think JE is good just not right away. Once you have a good grasp of Japanese, going JE can help cement your learning even further. Its not that its bad, just bad for a beginner, as they don't have enough experience to process all that information.

you mean EJ right?

amtrack Wrote:

Aspiring Wrote:If we list out your main points, your argument seems very biased. :O
Most people [strike] on this forum[/strike] would easily disagree with your viewpoint.

In support of your argument, I'll say that it is always best to learn and understand something the first time you encounter it.

Biased in what fashion? I don't recall showing any bias towards anything. The SRS claims just remind me a lot of things like the "shake weight". Sure some scientists say this works, but it is in no way widely endorsed by the scientific community.

And if I had all the room to type in the world, I'd tell you why I agree with you. I was heading that direction, but it'd take me too much space.

oefirouz Wrote:Spacing does do more to strengthen memory than cramming, which is why an hour a day over 10 days lets you learn much more than ten hours in one day.

I am no proponent of SRSing everything, but the SPACING EFFECT (not the software, the phenomenom) is backed up strongly in science and I would encourage everyone to ignore any advice you have on the topic.

Firstly, nowhere did I mention cramming. I thought it was mostly common sense that you cannot force yourself to learn something, which is the definition of cramming. Two, there is nothing I said that is incorrect. Examine your memory and tell me I am wrong. Do it, I dare you.

I can bet you several things. You know less from "spacing" than you do from constant exposure. Two, someone actively immersed in a language can learn more than they would by "spacing" out a limited exposure. Three, you can change the spacing in SRS at your leisure and it would still work for you. Four, you can review early and achieve the same or better in terms of results. Five, despite spacing, there are some items you just can't seem to remember.

The 5 points above illustrate my main argument: artificially spacing content does not strengthen your memory. It "maintains" your memory..if you space it just right (impossible btw), but it certainly doesn't make it stronger. If we apply a little bit of reason its not hard to come to this conclusion.

You have a variety of things that happened merely once, but you will never forget. Why is that? You have a variety of words you read or heard only once, but again, will never forget. Why is this? This alone proves memory is certainly not a product of some spacing algorithm. Any psychologist could tell you that your strongest memories have absolutely nothing to do with spacing.

EDIT: There are ways to strengthen your memory using SRS, but you have to use it properly.

Affirming a disjunct, argument from (a false) silence, kettle logic, and so on....
None of your arguments are in anyway valid in that there's credible amounts of scientific research that supports the SRS algorithm. Coming from someone who is entirely neutral (I'm not Woznik, I'm not payed, I don't code an SRS app): The SRS algorithm is significantly superior in just about every way despite your confirmation bias, and you cannot do anything about this claim until you can cite scientific evidence that contradicts the scientific evidence supporting the SRS algorithm.

amtrack Wrote:Sure some scientists say this works, but it is in no way widely endorsed by the scientific community.

This is a huge fallacy, and not a proper argument. The fact remains that credible scientific evidence in support of the SRS exists whether this is "widely endorsed" or not, in fact the Theory of Relativity was not widely endorsed at first either. What is it now? The cornerstone of modern physics.

amtrack Wrote:You have a variety of things that happened merely once, but you will never forget.

Show me one facet of information that you've only experienced once in your life and will never forget; things and "experiences" with emotional impact (IE: I lost my virginity and that only happened once, so the SRS must be false!) cannot be equated to the retention of dry facts.

amtrack Wrote:I can bet you several things. You know less from "spacing" than you do from constant exposure.

Spacing is an asset the SRS algorithm borrows that gives you the minimum amount of exposure needed to retain something, therefore it's impossible for you to distinguish (in the context of you requesting readers to look back randomly at their life, as one would need logs) what you "know" from just from spacing, and "know" just from constant exposure; at a minimum your "constant exposure" still consists of very small, albeit, unnecessary intervals.

amtrack Wrote:Two, someone actively immersed in a language can learn more than they would by "spacing" out a limited exposure.

A language is a massive amount of information, however diminishing returns exist; once someone knows enough then "constant exposure" to a majority of the language, in the context of memory strength, is wasted time in that most normal people have no need to reinforce simple words in their native tongue unless you assume an unreasonable gap of non-exposure — the existence of these diminishing returns logically supports the idea that SRS learning is more efficient than "constant exposure", and the reason "constant exposure" is advocated in the language learning community is that it (imitating a native) is very effective when you suck. Yes constant exposure happens quite naturally for anyone using the language, but it's not unreasonable to assume that for any fluent adult this exposure, for the sake of memory strength, is unnecessary in that a fluent speaker could take a year break and still retain a good 90% of his vocabulary.

Even AJATT, the place called "All Japanese All The Time", later modified his theories from "Contact Volume → Critical Mass → Ownage" to "Proper Frequency → Ownage ↔ Maintenance."

This, along with all scientific evidence that directly supports the SRS algorithm, is overwhelming proof of the superior efficiency and saved time of the spacing effect.

amtrack Wrote:Three, you can change the spacing in SRS at your leisure and it would still work for you.

[citation missing]

Anecdotal evidence is not proper evidence. If you can prove a correlation between unconventional SRS algorithm tweaks (in the sense of not recommended for optimal retention) and memory strength then by all means be my guest, however in the meantime do not make ridiculous claims without proper evidence.

amtrack Wrote:Four, you can review early and achieve the same or better in terms of results.

[citation missing]

Despite no credible source for what you're suggesting there are also two flaws with this premise in that even if you review early you aren't addressing the fact that you could be wasting time, and also that I don't think, even in your anecdotes, that you've kept long-term logs to truly know which has given you better results; you just kind of "think" it's better because it "always worked for you". Either way you would still need to prove a correlation for this point to be accepted.

amtrack Wrote:Five, despite spacing, there are some items you just can't seem to remember.

There is a random aspect to memory, and this is why people occasionally forget words in their native tongue that occupy the top 500 in frequency. The SRS algorithm has never claimed to "solve" all aspects of memory, so such a ridiculous logic from you isn't a proper argument.

amtrack Wrote:The 5 points above illustrate my main argument...

If anything these "5 points" work against the credibility of your argument in that they're all logically flawed, or baseless, or both.

Common trap:
------
OMG, someone is wrong on the Internet. I have to...
------

^^lol
Obviously both.
You fell into the trap. :O


Anyways... I downloaded corePlus.
Mighty Morphman allows you to enter any text into your knowledge database. I'd read articles and books, enter them into Morphman, and that would reorder core.

I also use yomicchan, but ^that's a nice alternative, considering core only has words with the highest frequency.

I'd sprint through the new words which I mostly know, and then continue reading or watching.

(srs is best for reviewing and storing knowledge....)

ryuudou Wrote:If anything these "5 points" work against the credibility of your argument in that they're all logically flawed, or baseless, or both.

I think you're misunderstanding several things.

1. By spacing I'm referring very specifically to the SRS algorithm. Obviously spacing is natural in that NO HUMAN CAN POSSIBLY STUDY ALL INFORMATION 24/7. I thought it was obvious. Unless you're just trying to be argumentative.

2. Just because some science supports it, doesn't mean it can be accepted as true. There's a reason the scientific method requires an experimented to be repeated as many times as possible, with as many different scientists as possible. It is a theory until proven otherwise, and a theory is not necessarily true.

3. People learn and continue to learn without an SRS algorithm. Thats all that needs to be said about it.

4. The spacing required varies from fact to fact. Some things you only have to see once in a blue moon. Some things you need to see more often. There is no one-size-fits all algorithm. If you want proof, tell me you have 100% retention rate in your SRS. You do NOT.

5. Again, any psychologist will tell you that your strongest memories have little to do with arbitrary spacing between events. You can dismiss it if you want, just because its an inconvenient truth, but it IS in fact true. It is not as simple as "spacing".


If you want to believe that SRS is the greatest way to learn/memorize anything that is fine. But for one, there is not enough science to support this as fact. Secondly, it is a *theory*, and theories are not necessarily true. Finally, you cannot dismiss my skepticism, as a large group of people smarter than I don't seem to backing up those "claims".

In other words, my opinion is fair, and its certainly not BS. If you require further proof of this, please read how an SRS is supposed to be properly used. You'll find that in no suggested method of study will you find a simple "just use it". Most suggest sentences, repeating things in various contexts, and the like. This is because the spacing algorithm by itself *cannot* strengthen your memory. Using it with the right tools can, but it by itself cannot.

Hyperborea Wrote:

amtrack Wrote:2. Just because some science supports it, doesn't mean it can be accepted as true. There's a reason the scientific method requires an experimented to be repeated as many times as possible, with as many different scientists as possible. It is a theory until proven otherwise, and a theory is not necessarily true.

Ummm, I don't really have a dog in this fight but I am a bit distraught by the twisting of the basis of the scientific knowledge.......

....If you have some science that disproves the claims that are made by the other poster then please post them.

He doesn't. His entire response was more or less all fluff if you noticed.

amtrack Wrote:2. Just because some science supports it, doesn't mean it can be accepted as true. There's a reason the scientific method requires an experimented to be repeated as many times as possible, with as many different scientists as possible. It is a theory until proven otherwise, and a theory is not necessarily true.

Like Hyperborea, I have "no dog in this fight," (a lovely phrase). And he is correct in that no theory is ever proven.

Generally the grounds for rejecting a theory is that there is a better theory that better explains the results on experiments. Until then the theory that best supports the facts in evidence is taken as provisionally true.

Is there a better theory that better explains the data? I'm not really familiar with the field of memory research.

CJ

@amtrack
There is evidence to support that the algorithm itself doesn't matter, and that even rough spacing is sufficient to produce the same effect.

Yes, spacing is a tool to be used in conjunction with mnemonics, feedback, input, and active practice.

The main point in this entire discussion is not the question of SRSing or not, but rather whether the usual spacing is optimal or not.
Somewhere in this threat someone said he'd increase the default spacing by something like 2000%. I use Anki 2. It is only possible to increase the spacing by 999% through the interval modifier. I tried this right from the time this was proposed, maybe 2 or 3 weeks ago.
As a consequence my total review time decreased by about 2/3rd, my correctness percentages didn't change (usually around 90%, +-5%), and I kept the number of new cards unchanged compared to before.
Granted, 2 or 3 weeks is too short a time to judge the real long-term impact of this change, but so far I am happy with it. I can use the time gained for reading now.

matsu Wrote:The main point in this entire discussion is not the question of SRSing or not, but rather whether the usual spacing is optimal or not.
Somewhere in this threat someone said he'd increase the default spacing by something like 2000%. I use Anki 2. It is only possible to increase the spacing by 999% through the interval modifier. I tried this right from the time this was proposed, maybe 2 or 3 weeks ago.
As a consequence my total review time decreased by about 2/3rd, my correctness percentages didn't change (usually around 90%, +-5%), and I kept the number of new cards unchanged compared to before.
Granted, 2 or 3 weeks is too short a time to judge the real long-term impact of this change, but so far I am happy with it. I can use the time gained for reading now.

Since 999% is a 10 fold increase, and since 3 weeks is 21 days...

Unless your intervals were on the order of 2.1 days you wouldn't have even seen the results yet.

-_-

I have set it to 200 without any noticeable consequences, except that my reviews quickly went down to ~100 per day to ~70, but now they're back to ~90 again, because I encreased the number of cards added per day.

To me, the spacing is helpful because it limits the number of flashcards I have to review every day. I only review about 100 of my 3000 or so cards each day, and my 90% retention seems to "prove" that it works for me.
It does require a lot of immersion to work proberly though, and I never "learn" anything during my anki reviews.

100/3000 cards??
:O

I leave the interval modifier at 100%..
But, depending on the deck, I set the easy bonus between 200-600%.

I add about 10-20 a day; don't want the reviews to go out of hand.
Mostly (monolingual) subs2srs + "definition branching" cards nowadays.

I have an extra 20 day button instead (there's a plugin) to instantly make it mature if the word is so easy that I really shouldn't have added it to begin with. I have it on 100% on the RtK deck though, and 200% an both my Japanse and German sentence decks.

I also have a 25% spacing on forgotten cards, so there aren't many "relearning" cards either.

I think some language learners are skeptical of SRS due to the fact that the effects of over-learning does not play into their huge language decks. SRS really shines when you combine the "spacing effect" with the effects of over-learning. I have been keeping my acrostic memory system statistics everyday for the last couple of months. You can clearly see the spacing effect in action. I broke the learning stages down on my page.

http://www.blurb.com/b/4121439-math-engl...-acrostics
http://www.fileden.com/files/2012/8/25/3...-21-40.png
http://www.fileden.com/files/2012/8/25/3...-12-05.png

"amtrack" If you are still skeptical here are all of the statistics.
http://www.fileden.com/files/2012/8/25/3...0stats.zip

Wow. amtrack's thinking is almost as dated as the crummy trains company he runs.