Learning mnemonic techniques is recommended
(Eli Liang, Russia, Oct 03, 2001)
I am given pending items only once to memorize and then the system moves me on to the next pending item. This is not enough time to really memorize
You will be asked all difficult items again within Final Drill. You need to strive at being able to see an item once and remember it well (in a given learning session on a given day). This is a question of mnemonic skills that go beyond the scope of SuperMemo. Hopefully, after a few months, those skills will come naturally. It is also recommended that you go to sites that describe various mnemonic techniques to experiment if any of these could apply in your case
You do not have to
interrupt repetitions to introduce corrections
(Deron Isaac, USA, May 21, 1997)
How can I edit texts of items during repetitions without backing out of the learning mode?
Choose Q to edit the question, A to edit the answer, E to edit all text components or Ctrl+E to switch all components to the editing mode. You can also edit component properties by using the component menu available with a right click over a component
You can view the calendar
of repetitions with Ctrl+W
(Matt Cassidy, New Zealand, Aug 18, 1997)
How can I list all elements that are to be repeated on a given day in the future?
Use Ctrl+W (Tools : Workload) and double-click the day of interest
Grades provided in final
drill do not affect learning
(Grzegorz Malewski, Poland, Dec 10, 1997)
Do grades at final drill affect the learning process?
No. They are only used to eliminate items from the final drill queue. All items with grades above Pass will be eliminated
You can use Forget
to postpone memorizing a given element
How can I postpone memorizing an element when I am learning new material?
Choose Forget on the element menu. This will shift the element to the end of the pending queue. You will learn the element only after you memorize the entire pending queue (unless you sort the pending queue or memorize the element manually)
SuperMemo has tools enabling you to effectively deal with the backlog accumulated during vacations
I left for vacation, and my learning process is in mess. What should I do?
Check Learn : Postpone : Auto-postpone to automatically postpone repetitions left from the previous days at a beginning of a learning day. Alternatively, use Learn : Postpone : All or Tools : Mercy
You can press F5 to see learning parameters
(Eric Chen, Japan, Fri, Jun 22, 2001 3:06)
How can I easily see the learning parameters of an element?
You can press F5 to see all the most important statistics. You can preserve such layout with Ctrl+Shift+F5
to permanently remove elements from the learning process
(Len, USA, Feb 11, 2001)
When I run through the learning process, SuperMemo keeps showing me topics, concept elements, as well as empty elements. So in between questions I see a blank page, and the title bar says Asian Geography or Reading List. It seems very silly
You need to specifically inform SuperMemo which element should and which should not be included in the learning process. By default, all new topics and items are memorized. To permanently remove an element from the learning process, choose Learning : Dismiss from the element menu or press Ctrl+D. To introduce an element back to repetitions, choose Learning : Remember or press Ctrl+M
You can locate most difficult items
(Noel Clary, USA, Aug 17, 1998)
I have created my own database on plumbing and air conditioning. My forgetting index is quite high. Are there any tools in SuperMemo which could help me remedy this situation?
You might want to use View : Other : Leeches (Shift+F3) and locate the elements that cause most problems in learning. You must then go into your own mind to answer the question why these elements are hard to recall. Usually these are too complex, too long, too boring or too similar to other elements in the same collection. You can also send 3-4 most difficult items to us for review to receive some suggestions. Read more: Leeches. You could also have a look at: 20 rules of formulating knowledge in SuperMemo
You do not have to interrupt repetitions to forget or
dismiss an element
(Josef Holubik, Czech Republic, Sep 13, 1998)
When I make repetitions, I would sometimes like to use Forget or Remember but I only have grade buttons available. Do I have to stop the repetition cycle?
No. You can use Learning : Forget from the element menu and Ctrl+M for Remember. In both cases, you will continue with the next item in the repetition schedule as soon as resetting or recommitting an item is completed
Completing the final drill is not essential
(David Mckenzie, New Zealand, Sep 3, 1998)
What should you do if you don't complete the final drill for the day? Thoughts are: 1. Delete it, 2. Complete it first thing next day, 3. Complete it the next day, but with other final drill items. (I usually do 3.)
It would be best to go through the final drill every day. Second best would probably be to delete final drill (Learn : Cut drills). This comes from the fact that all repetitions made on dates different from the original schedule interfere with computations made by SuperMemo. You can simply fool SuperMemo into believing that you know an element better than it is indicated by the history of repetitions. Finally, it would be worst to go through final drill before repetitions on the next day (for the same reason as above). The picture gets more complicated when your first interval for forgotten items is relatively long (e.g. over 5 days). In these circumstances doing the final drill on the next day might be more profitable than not doing it at all
You can always delete the final drill
I noticed that File : Repair collection does not recover the final drill queue. What if drill.dat file becomes damaged? Will that make learning impossible?
No. drill.dat is a temporary file and can always be removed with Learn : Cut drills with little damage to the learning process. This file can also be deleted manually
There is learning material available for download
(Roger Dykes, USA, May 20, 1998)
What ready-made learning material is available there for SuperMemo?
See SuperMemo Library.
Note that SuperMemo 17 can only upgrade collections in SuperMemo 16 format. However, all collections for SuperMemo 8, SuperMemo 98/ 99, SuperMemo 2000/2002/2004/2006/2008 or SuperMemo 15 can be upgraded to SuperMemo 16 by simply opening them with File : Open collection.
You can easily repeat sounds, display translation or
(Tomas Klinkovsky, Czech Republic, Aug 7, 1998)
How can I do the following actions using the keyboard in Advanced English:
You can repeat sounds with Ctrl+F10. You can also display translations with Alt+U and the phonetic transcription with Alt+Y
You can remove all information about the learning
(Mariusz Szepietowski, Poland, Dec 7, 1997)
How can I remove information about the learning process from a collection? I would like to start learning all over again from scratch?
Use File : Tools : Reset collection.
You can backup learning in large collections
(Michal Grodzki, Poland, Aug 20, 1998)
You advice the users of Advanced English to make back-up copies with File : Export : Learning process. However, this does not back up the material introduced by the user. What is the best solution in a case when I want to add lots of my own items to Advanced English?
You might try one of the following:
Knowledge tree does not play a role in learning
(Nita L. Wunderlich, USA, Aug 24, 1998)
How can I make a good use of the knowledge tree in the learning process?
Structuring knowledge by means of the knowledge tree is not very important for the learning process itself; however, it greatly simplifies managing large bodies of facts pertaining to different subjects. The knowledge tree is best managed by means of knowledge groups. Read more: Using concepts
You can view all pending elements in a given branch
(Shaun Hoffland, UK, May 25, 1997)
How can I view all pending or dismissed elements in a given branch in the knowledge tree?
Probably the simplest method is to:
After the above sequence of operations, the browser will contain only the pending or dismissed elements belonging to the selected branch
Show answer stage may be skipped
(Anatolyi Lipatov, Ukraine, Aug 28, 1998)
When I add new elements in the contents window, it happens that the repetition cycle skips showing the answer even if the answer component is marked as Answer. Is it a bug?
No. This is a standard behavior for topics. You must check Type : Item on the element menu. Topics are used in passive overview or reading. They do not respect the Answer attribute on the component menu. They are just shown as they are and expect you to click Next repetition (i.e. skip Show answer and grading) (see also: Topics vs. items)
You can quickly make repetitions in a selected subset
of the learning material
(Mike Condron, Saturday, January 06, 2001 5:22 PM)
Is it possible to make all outstanding repetitions related to a narrow subject; for example to the abdominal aorta? Say I have an exam tomorrow and not enough time to make all outstanding repetitions
Yes. Search for "abdominal aorta" (e.g. with Ctrl+F) and choose Learning : Learn on the browser menu (e.g. with Ctrl+L)
You can make repetitions in a single branch
(Patrik Nilsson, Tue, Aug 21, 2001 11:24)
Is it possible to have different priorities of items in the pending queue? E.g. if I first want to memorize what about school and then all kinds of hobbies?
Pending queue is linear but can be sorted by various criteria. However, the recommended solution in your case is to keep school and hobby material in different branches or different concept groups. Then you will be able to use branch learning. For example, select School branch in the contents window and click Learn (or choose Ctrl+L)
If you disregard the grading system, your learning
progress will be negligible (#3762)
(Michal Ryszard Wojcik, Poland, Tue, Feb 20, 2001 13:45)
I've been coasting along with my German collection. I gave a 5 for a fully correct response and a 4 for a mistaken response. I was taking it easy. I'm going to keep using my German collection like that. It does not burden me and it keeps my German in memory
You could spend your time better by just watching some German TV channel. This method will have negligible effect on your knowledge. Grade Bright (5) will keep the easy core of your knowledge in memory. Grade Good (4) will produce short-lived refreshments with a negligible long-term effect. This is nearly the exact statistical replica of your encounters with knowledge in traditional language learning assuming your collection is representative of the language. This means that there is no special treatment of forgotten pieces. In other words, this is not the SuperMemo method and using SuperMemo here verges on a waste of time. You could read some good German book in this time and kill two birds with one stone (learning the language and learning the subject matter)
Mnemonic techniques should be used on items that are
difficult to remember
(Adam, Australia, Sunday, August 05, 2001 3:26 AM)
I have 673 items in my Chinese-English-Chinese database. Some items are very hard. The Leech dialog box comes up over and over again, and I do not know how to get rid of it. I want to keep difficult items in the database. Should I reset the items, i.e. send them back to the pending queue?
Chinese is inherently difficult to a European or to an American. Some items may comply with 20 rules of formulating knowledge and still fall into the leech category. The only solution is mnemonic techniques. Please have a look at mnemonic websites to look for techniques that would best match this kind of material. A popular way of learning ideograms is to build pictures nested in their skeleton. If pictures are semantically related to the English word, the recall may become far easier. With time, you will develop a subconscious skill of visualizing ideograms in your own way with benefits to recall
Use Reschedule on a branch to progressively introduce
new elements to the learning process
(Justin Wilson, USA Educational, Monday, August 06, 2001 3:39 AM)
I wish to have certain branches progressively added to the learning process, at a certain specified rate such as 2 items/day. I have tried the Spread option, but it seems it did not do what I wanted because today my learning process had 90 items in the Final Drill as opposed to the normal 10-15
Using Spread is still probably the best way to accomplish your goal. If you use Learn : Cut drills, you will eliminate the final drill queue with little damage to the learning process. The only major shortcoming of using Spread is that many elements will enter the learning process with the first interval much longer than usual. However, it should not affect the learning process much if your branches are of similar difficulty as the rest of the material, or if they do not make a large proportion of memorized items
Wrong recall may indicate memory interference
(Mike Condron, Tuesday, September 18, 2001 2:39 AM)
Why no recall is scored less than a wrong recall? Suppose I'm trying to remember "What's the capital of Canada?" and I answer "Warsaw." This seemed to me to be a much worse state of affairs than "I don't know"
Wrong answer may put you in more trouble or shame in life, but we are interested in the state of memory here. Wrong recall is usually caused by interference from other memories and is less indicative of lost memory traces. Zero recall, on the other hand, is a clear indication of lost or severely disrupted memories. Your grades should penalize you solely for poor memory performance. However, you can use priorities, the forgetting index, Learning : Reschedule on the element menu, or other tools to ensure that mission-critical pieces of knowledge stay permanently etched in your memory. In other words, you should never grade yourself more harshly on important material as this will fool SuperMemo as to your performance
Resetting the measured
(Peter Cool, The Netherlands, Nov 6, 1998)
I started with SuperMemo 70 days ago (your French plus some words added by me; total 1000 words). In the first weeks I made a lot of mistakes so my measured forgetting index was 20%. Although I make very few mistakes now during repetitions the forgetting index decreases very slowly. Is this normal?
The measured forgetting index includes the record of all repetitions made since you started learning. That is why it changes at an ever decreasing rate. If your performance is good and you would like to more accurately check your current forgetting index, you might reset the forgetting index measurements with Tools : Statistics : Reset parameters : Forgetting index record. It will not affect the learning process per se
Forgetting index vs.
(Tomasz Szynalski, Poland, Oct 18, 1998)
What value of the forgetting index ensures the optimum ratio of (retention)/(time spent per day)?
Paradoxically, the highest speed of learning can be accomplished ... without SuperMemo! In our daily life we pick up lots of facts that stay in our memory for long with few repetitions in lifetime! The problem is that these are usually not exactly the facts or rules that are critical to our goals. In other words, not the speed of acquiring new items counts but the speed of acquiring new items bearing a given content.
It is difficult to determine exactly what forgetting index brings the highest acquisition rate. Simulation experiments have consistently pointed to the value of 25-30%. You can even plot speed-vs.-forgetting graph using your own actual learning material in SuperMemo using Tools : Statistics : Simulation. You will probably also arrive to similar results
As you perhaps know, SuperMemo disallows of the forgetting index above 20%. This comes from the fact that you should aim at achieving high speed of learning combined with high retention of the learned material. Setting the forgetting index above 20% would be like giving up SuperMemo altogether and coming back to remembering only that what is easy to remember. In highly interlinked material where new knowledge depends on the previously acquired knowledge, high forgetting rate can even be more harmful
Nevertheless, if you want to maximize the speed of learning with little control over what actually stays in your memory, set the forgetting index to 20%
Learning in portions
(Robert Szumilo, Poland, Jan 3, 1999)
What is the optimum approach to making repetitions with SuperMemo: one long session or a few smaller sessions (e.g. main repetitions in the morning and the final drill in the evening)?
For psychological reasons, the quality of learning should increase substantially when working in separate sessions, esp. if the number of repetitions surpasses 100 per day. Additionally, a break before final drill is useful due to the spacing effect (more exactly: the lag effect here). The danger of this approach is ... you can easily drive yourself into a situation in which you will spend excessive proportion of your day on repetitions (in the future when your schedule changes you might have problems with keeping up with your present pace)
(Christian Roessel, Germany, Jan 10, 2001)
I found items that have fewer repetitions than lapses. What is the interpretation of Repetitions?
Repetitions in the element data window do not show the actual number of repetitions of the displayed element. Repetitions display the number of repetitions since the last time the element was forgotten (it includes the use of Forget). To get the actual number of repetitions you can inspect the repetition history with Ctrl+Shift+H.
(Manfred Kremer, Germany, Sep 7, 1998)
I noticed that frequently I get Optimum Interval in Element Data window shorter than the last interval displayed as Interval. Is it a bug in SuperMemo?
No. If your forgetting index is very low, e.g. 3%, SuperMemo will often conclude that you will stand 97% chance of remembering a given element only if your next interval is shorter than the presently used one. In such cases, it will not accept the new value and the new interval will be at least 5% longer than the previous interval. Please note that the forgetting index equal 3% should only be used for selected high-priority items. Keeping the forgetting index at this level throughout the collection will make repetitions annoyingly frequent and ineffective
To avoid the rigid repetition schedule ignore the
(dansujp, Sun, Sep 16, 2001 3:07 PM)
I dislike the fact that SuperMemo forces a repetition schedule on the user.
The schedule is a result of the algorithm that takes retention criteria as guidance. As a result, it is not SuperMemo that determines the schedule but the state of the student's memory. The simplest way of living with this rigid plan for repetition is to stop looking at the Outstanding field. If you ignore the statistics, SuperMemo will let you make your repetitions at irregular intervals. If you come too late, it will simply not schedule new material, and reduced retention will generate more workload
You can sort repetitions
(P.M., Aug 28, 2001)
How can I sort repetitions from low to high intervals?
You can sort your repetitions by the length of the interval using the following method:
SuperMemo cannot change the date of the last
(Patrik Nilsson, Thursday, September 20, 2001 1:18 PM)
Is there anyway of retaining the repetition interval when doing Mercy?
No. Intervals are defined as the period between the last repetition and the time when the next repetition should be scheduled. You cannot change the former (otherwise you would fool SuperMemo as to how strong your memories are). If you change the scheduled day of repetition, the interval will change accordingly
Setting a queue of branches for learning
(M—, Poland, Wednesday, October 11, 2000 7:57 PM)
I need an option "Set branch learning queue". For example: at the beginning repeat branch X, then Y, and Z, etc.
You can create a subset with those branches in sequence. Then you can choose View : Subset and Learn (Ctrl+L). However, if you add new elements to one of the branch, you will also need to insert them into the subset
You can learn and review your exam subject (#908)
(CMaggio99, Thursday, May 02, 2002 11:08 AM)
I would like to be able to test myself on one particular topic or set of topics (i.e., lecture notes) when an exam is coming up. How would I do this without having to learn non exam pertinent material?
You have several options. If you start learning early enough (e.g. several months in advance), it may appear that you do not overload the learning process and can retain all material in memory with the programmed retention. Otherwise, you could do OR-search to provide a search definition for your particular exam and use Review (on the Process browser menu). You could also save all material belonging to a given exam in a subset. You could also keep separate exams in separate branches of the knowledge tree. Always use Review (mid-interval repetition) on your branch, subset or search definition to ensure you maximize your retention (even beyond the retention programmed in SuperMemo).
You can sort items by intervals
(E.G., Poland, Monday, October 09, 2000 12:33 AM)
How can I sort items from long to short intervals?
Randomizing repetitions in a subset
(Forum, May 24, 2002)
Is there a way to randomize your learning within a subset?
You can execute outstanding repetitions in a subset in a random sequence:
Use Ctrl+M to introduce elements to the learning
(John McGrath, Netherlands, Wed, May 29, 2002 10:35)
I have just purchased the Dutch collection. The problem is that I can not work out how to select which items from the 2,500 items should be first introduced to learning
You can use navigation (e.g. the contents window, browser, etc.) or search tools (Ctrl+F) to locate items of interest and then choose Remember (Ctrl+M) to introduce them into the learning process. You can also use Learning : Remember in the browser to introduce many elements at once. For example, search for a difficult word with (Ctrl+F) and memorize all items that use this word with Learning : Remember on the Process Browser menu
SuperMemo leaves the memorizing rate in your hands
(Patrik Nilsson, Tuesday, October 02, 2001 5:56 PM)
Is it possible to set the first repetition to start in 14 days and a new element appear every second day?
No. SuperMemo does not provide tools for scheduling pending items. You have to choose Learn : New material manually on the 14th day and repeat that procedure every second day. Alternatively, you could introduce all these elements into the learning process, change the date to the 14th day and reschedule the branch in question with an average of one element every second day. In this case, all elements that score well in the first repetition will be sent to remote intervals in accordance with the spaced repetition algorithm
You can copy your learning material or your learning
process from one computer to another (#7293)
(Jarowiecki, Pawel, Poland, Thu, Oct 11, 2001 15:20)
How can I transfer my learning status from one machine to another?
The simplest method is to transfer the entire collection with File : Copy collection. If you do not edit your learning material on one of the computers (i.e. use it for learning only), you could also use File : Export : Learning process (to export your learning progress from one computer) and then File : Import : Learning process (to import it back on the other computer)
You should always keep your date correct in your
(anonymous, Sat, Oct 20, 2001 1:49 AM)
Is it ok to play with the system clock behind SuperMemo's back?
No! SuperMemo optimizes repetition dates using the record of retention in time. If you change the date in your computer, you will not only make your repetitions be scheduled sub-optimally, but you will also damage the data model that will affect all future repetitions. You should pay utmost attention to making sure your system date is correct. If you want to reschedule outstanding material, use rescheduling tools such as Postpone, Mercy or Dilute
Postpone does not conflict with the idea of SuperMemo
(Beta, Maxim, Poland, Mon, Feb 04, 2002 20:04)
I see that you invested a lot of time in Postpone options. However, what is the sense of using Postpone if this increases intervals beyond the optimum?
Postpone makes it possible to bridge pure SuperMemo with traditional HIHO learning (high-in, high-out, i.e. lots of reading, lots of forgetting). You can start your day with creative reading, browser, studying, learning material in all interesting fields. Then using Postpone, you can narrow this to strategic knowledge in your professional field. Then you can Postpone all topics and focus only on repetitions of the old material. Towards the day's end, you can focus entirely on mission-critical knowledge (e.g. as identified by low requested forgetting index). Your overall forgetting rate will be naturally higher; however, your core material will retain high levels of knowledge retention. Instead of having 200 elements scheduled for repetition, you may get 4000 of which a large portion will only be reviewed stochastically. Only the core 50-100 elements per day will be repeated with full adherence to pure spaced repetition
New elements are memorized automatically
(Beta, Fri, Mar 22, 2002 13:37)
Why does SuperMemo automatically memorize all items I add with Add New?
The learning process in SuperMemo 2002 and later is by far more efficient an optimization tool than the linear pending queue of the earlier versions. Memorizing all new material occurs by default and is the recommended course of action. You can remove elements from the learning process with Learning : Forget from the element menu or by means of the subset and/or branch operations. This new default behavior reflects the new strengths of the program. The pending queue might be used only as a reserve tool for material that can be memorized with substantial delay
You can react to items forgotten in mid-process
(Beta, Feb 07, 2002)
When I review some branches, I often find items that I have forgotten. In the past, all I could do was to rememorize them. Can SuperMemo handle it in a more intelligent way?
Yes. You can execute a mid-interval repetition (Ctrl+Shift+R) with a failing grade. This will let SuperMemo correctly interpret that status of the forgotten item, upgrade its difficulty and modify learning optimization parameters accordingly. You can also use this procedure to score a passing grade on items you know well. SuperMemo will delay the next repetition accordingly
Postpone simulation will not be replicated at Postpone
(Beta, Sweden, Feb 09, 2002)
I simulated Postpone and got Average delay interval: 58 days. Then I ran Postpone and got 62 days instead
Postpone is partly stochastic due to random interval dispersion (as in standard repetitions). For that reason, your results will always differ slightly in each simulation run and in the actual Postpone
Browsing past repetitions is inclusive of elements
that have been dismissed or made pending
(Beta, Feb 06, 2002)
Why I have 10 topic reviews on a given day in statistics, while the browser opens 11 topics on that day?
The workload statistic includes only repetitions and topic reviews. The browser also opens elements that have been dismissed on that particular day (i.e. not necessarily reviewed). In your case, you might have had 10 topics reviewed and one topic dismissed (e.g. after search)
Checking the outstanding material
(Beta, Romania, Fri, Mar 22, 2002 13:37)
The disabled Learn button could be a very useful indicator of Outstanding=0+0+0. Now, every day I must choose Ctrl+W to view what is the situation in my collection
The recommended way of inspecting the number of outstanding elements on a given day is via the status bar which can be saved in the displayed state in the default layout. In addition, for higher screen resolutions, the Statistics window is the best quick-glance tool for inspecting the statistics. It can also be saved in the default layout (Ctrl+Shift+F5). Moreover, it is rather a rare occurrence to have no outstanding material for a selected day. Even 2-3 minutes of work per day will ensure there are some repetitions in the process on the assumption you work regularly and keep your knowledge in a single collection. Single body of knowledge is highly recommended, esp. if your learning is not intense. The greater the learning sample, the better the optimization work done by SuperMemo in scheduling repetitions
Daily and Monthly workload averages
(Beta, Sean Toner, United Kingdom, 25/03/2002)
In Tools : Workload, I have the totals of 37 for Mar 27, 46 for Mar 28, 53 for Mar 29, 49 for Mar 30, and 47 for Mar 31. If I press the Monthly tab, 7.5 is displayed for March. This number appears meaningless
Use Daily Workload to see the average of items scheduled for the remaining days of the given month. Use Monthly Workload to see averages for whole months
SuperMemo may ask you questions you have never seen
(Jon, Thu, May 30, 2002 20:12)
I have bought several collections. When viewing for the first time, should I just click Learn? Something doesn't seem quite right about this because I have to guess the answer and rate it before I have even seen the answers at all
You are right that in your case immediate viewing of the answer would be equally effective and even faster. However, usually there is a proportion of material the student already knows. In such cases, using Show answer and grading helps eliminate items from the first final drill. If you do not know the answer, you can press Show answer and grade yourself Null on all items. Naturally, you should still pause and try to remember. In the early days of SuperMemo, users usually created their own question-answer material. This behavior has been inherited from those days. There is no simple way of pointing which elements are already known. If you know none of them, you could, theoretically, execute Remember in the browser, but again, you will be shown question during the first repetition. At that time, SuperMemo will make the first determination of difficulty estimation, which makes sense only if you have been exposed to the material while memorizing it. In short: use Learn to memorize material even if you know no answers
SuperMemo adjust intervals to the selected retention
(Ole F., Germany, Tue, Jun 04, 2002 17:06)
When I append a new word and learn it immediately, I want to be asked this word the next day
You will need to set the interval manually (e.g. with Ctrl+J). SuperMemo will always choose an interval that complies with your target retention level. Choosing shorter intervals slows down learning and will often reduce retention in the long-run (due to the spacing effect). For those reasons, reducing interval manually is not recommended. You can do it, exceptionally, for vital pieces of information, for which, you can also reduce the forgetting index; however, remember that these operations will increase your workload substantially
Response time is never taken as longer than 30 seconds
(Jens, supermemo Quito, Ecuador, Friday, May 31, 2002 7:25 PM)
I wonder if the statistics are severely affected if I accidentally leave the computer while about to answer a question?
No. If you keep on thinking for longer than 30 seconds, SuperMemo will cut off the extra time. In no circumstances should you need to spend more than 30 seconds on an item. Long response time call for review of item formulation. Hence the cut off.
You can learn more details of past repetitions from
(Patrik Nilsson, Monday, December 24, 2001 11:04 AM)
How can I count the number of lapses on an element during its life (also before Forget, Dismiss, etc.)?
You can use Repetition History on the element menu (Ctrl+Shift+H).
You can turn Final drill off
(Arne-Johannes.Jenssen, Norway, Friday, August 23, 2002 1:17 PM)
Why no elements are added to final drill even if they were graded below Good?
Final drill can be manually turned off. Please check Tools : Options : Learning : Skip final drill if it is unchecked
You can define default element in SuperMemo
(Patrik Nilsson, Friday, November 16, 2001 9:54 PM)
One annoying thing is when you start SuperMemo you can see the answer of the next question. Please put a blank window up first
You can solve this problem by defining the default collection element with Tools : Options : Access : Default element
You can determine which element is presented when
(Shaun Hoffland, United Kingdom, Sunday, October 06, 2002 9:12 PM)
I do not like the idea that the first element that is presented upon SuperMemo start up is the element, which had last been viewed before SuperMemo was closed on the previous day. Is there a way to change it?
Use Tools : Options from the main menu, click the Access tab, and change the default element by clicking on the folder- tree button (to the right from the Default element text box)
"Final drill?" message may be suppressed
(Tomasz.Olszewski, Poland, Friday, October 18, 2002 2:29 PM)
When I finish my learning, I do not see "Final drill?" like in earlier versions
"Final drill?" message may be suppressed when you begin working with a new collection. Because this message used to confuse beginners, it is displayed only
"Nothing more to learn" shows up when all learning
queues are empty
(jude kendall, Wednesday, September 04, 2002 8:03 PM)
When I use Learn on the ABC collection I go through all the questions many times and I never get to the "Nothing more to learn" popup window.
"Nothing more to learn" shows up when all learning queues are empty. In standard learning (your case), this means that there must be no outstanding elements in the learning queue, no elements in the pending queue, and no elements scheduled for final drill. In subset learning, there must be no outstanding elements and no pending elements included in the learned subset (subset learning does not run through the final drill). As repetitions clean up both outstanding and pending queues, only the final drill queue can run indefinitely (i.e. elements taken from the front of the queue can be put back at queue's end). In your case, it is most likely that you simply never give a drill passing grade to one or more elements taking part in repetitions (e.g. you click a wrong answer in the multiple choice test). To test for this possibility, use Learn : Cut drills to make sure your final drill queue is empty. If you use drill passing grades (i.e. Good or Bright), you should eventually see "Nothing more to learn". The smaller your collection, the faster this will happen. Read also about grades and the repetition cycle
You can count lapses in a subset
(Patrik Nilsson, Tuesday, December 18, 2001 11:53 AM)
How can I see how many memory lapses occurred in a given subset of elements?
You could try two methods:
Medical Biology can be mastered within a year
(gobind, Monday, December 23, 2002 6:05 AM)
It is not understood how one would go about tackling an enormous collection - say Medical Bio Mix - and use the underlying scientific methodology of the SuperMemo program
10,000 items included in Medical Biology are not many as for SuperMemo standards. With one hour daily investment, you should be able to memorize the whole material within a year. The material is sorted by difficulty. This way you shall begin with simple basic knowledge that may already be familiar for you.
As there is no explanatory material, you would best combine learning with some incremental reading. If you encounter an unknown concept, e.g. "giardiasis", you can search the whole collection for "giardia", and introduce relevant elements to the learning process with Ctrl+M (Remember). If this does not help, you could search the net for some materials about Giardia and process them simultaneously with incremental reading. As you clear the initial obstacles, your learning process will accelerate and the material will no longer look intimidating.
As some portions of medical knowledge undergo constant change, you should also import additional articles to accompany items marked with a date. For example, if you get a question about the speed of sequencing DNA, you should check up-to-date sources as the progress in the field is lightening.
Important! If you ordered various pieces of Medical Biology mix separately, you can merge them all by using File : Merge collection. After this merger operation, you will yet need to sort the pending queue by ordinals. This is to makes sure that the pending queue (i.e. the order of learning) is sorted by difficulty
Optimum timing of the final drill
(Andy Y. Lin , USA Educational, Friday, October 03, 2003 4:21 PM)
When is it best to do the final drill? Right after I finish the regular scheduled repetitions, or wait until the end of the day?
It is very difficult to provide a solid algorithm. You can best try both methods for some time and see how it works for you. The problem is that you need to balance two forces:
on the other, later hours may make you less alert; as this is partially a defense against short-term memory overload, your efficiency may drop. As the alertness is strictly related to your circadian rhythm, the amount of sleep, the amount of prior learning, napping, etc. it is difficult to provide a prescription that would work for everyone.
The simplest formula would probably be: do the final drill as late as possible within your optimum mental effort hours.
Note that optimizing final drill is not likely to make a major effect on your overall learning results. You could then simplify the formula even further: do the drill when you like it most
Very long intervals
(Vinh L , Wednesday, February 19, 2003 12:26 AM)
Is there a way to shorten the intervals instead of lowering the forgetting index? The problem is that I have over 4500 items that use mnemonics, and 1000 items that don't use mnemonics. All these items are treated the same. The non-mnemonic items get too long intervals.
You can shorten the interval with Ctrl+J or Ctrl+Shift+R. The uniform treatment of easy and difficult items is a problem only at the first repetition. Once you provide failing grade on difficult items they will be considered "harder" and their intervals will shorten automatically. Before the first grade, however, SuperMemo has no way of knowing which items are difficult and all items get the same treatment. The length of the first interval will depend on the average difficulty of knowledge you learn. It can change over time. The longer the first interval the better! The long first interval is the evidence of your knowledge being well-structured on average. Unless you cannot "afford" forgetting and you need knowledge "right now", you should rather let SuperMemo handle the first repetition. Your first interval will be adjusted to your requested retention level (e.g. 95%). In your case, most of the forgotten items will naturally fall into the difficult category. If you start shortening the intervals on difficult items, you will improve the recall at first repetition, possibly increase the length of the first interval even further, and violate the principle of spaced repetition, which says that an early repetition may bring a better recall today but a poorer memory in the long run (due to the spacing effect)
Random Tests should not be part of your daily learning
(IngLand, Canada, Monday, February 18, 2002 7:35 PM)
Is my understanding correct that part of the theory supporting SuperMemo is that repetitions that are spaced to occur too frequently will actually interfere with the learning process of those items? In this regard, I am wondering whether it is advisable to avoid running the Random Tests feature on a frequent basis
You are right. Random Tests will interfere with the learning process. This is not primarily due to forming "weaker" memories, but mostly by fooling SuperMemo as to your true learning performance. If you get an item in a random test shortly before an actual repetition, you are likely to send the next repetition to a distant future. In addition, Random Tests are just a waste of your time. If you ask SuperMemo to keep a given level of knowledge retention and do your repetitions in time, you do not have to worry about the "safety" of your knowledge. Your time will better be invested in learning new material or other creative activities.
If you are facing an exam, it is far better to use mid-interval repetitions. This helps SuperMemo reflect the impact of the review with minimum damage to the learning process and minimum increase in the overall workload. See: Subset review
There is no optimum number of items to learn or
(Simon Jacques, Monday, March 04, 2002 10:20 AM)
I cannot find any information concerning the optimum number of items one should include for one repetition sitting or the optimum amount of time one should spend in one repetition sitting
A universal optimum number of daily repetitions cannot be determined. It strongly depends on many factors such as:
There is one guiding criterion though: enjoyability! If you like your repetitions, you can go on as long as you feel like. If you start feeling tired or impatient, you may only condition yourself to dislike SuperMemo. This can be the first step to never succeeding. Enjoyability is proportional to your ability to formulate knowledge, your mnemonic skills, your alertness, your ability to avoid stress, your love for learning, your hunger for knowledge, the applicability of knowledge, your understanding of SuperMemo and its limitations, etc.
Deciding between Pass and Good does not determine
your success in learning
(Ole F., Germany, Saturday, July 20, 2002 4:33 PM)
I know that an item that I judge Bad or Fail will be repeated sooner than one that I mark Good. But what about the items that I mark Pass?
The main boundary is between Pass and Fail. Pass means "I remember well enough". Fail means "I failed to remember to my standard". Good is simply slightly better than Pass. Good will also statistically produce slightly longer intervals. However, if you are not sure when to use Good and when to use Pass, you can use Pass then when you want an item to appear again in the final drill. The only place where grades are very important is where you distinguish between Pass or more, and Fail or less. Differences between Pass, Good and Bright are less important. Differences between Fail, Bad and Null are less important. Beginners can limit their grading to Good and Fail and still do well in learning!
Retention and Consolidation refer to the same set of
items tested for recall at different times
(Mike Condron, Wednesday, July 24, 2002 10:54 PM)
In the Workload screen, the option Retention sometimes shows a number, and sometimes something like "75.5->84.2". What does this mean?
In Workload : Retention, the single number or the first number in the pair denotes the retention as measured during repetitions. This is the percent of correct recall on this given day. The second number is consolidation. It shows the percent of correct recall on items repeated on this given day as measured in future repetitions. Consolidation is meaningless until the first recall of items repeated on the selected day is attempted. This is why it is not displayed until some time after the day selected in Workload. In your example, you remembered only 75.5% of the material on the inspected day. But later on, you were able to recall 84.2% of the material repeated on that same day. Both measurements refer to the same material
100% recall is unlikely
(d c bulldog, Monday, July 29, 2002 3:44 AM)
I'm beginning to try to learn the definitions to a long list of words, and I would like to remember them with 100% accuracy
With a growing sample of the learning material and over longer periods of time, 100% recall becomes a statistical improbability. For practical reasons and for very satisfactory results, you can aim at 95% recall (i.e. the default forgetting index of 10%). You can also increase recall by redundancy (i.e. formulating the same material in several ways) and proper formulation (e.g. mnemonic techniques), but both require lots of experience and patience. Default settings in SuperMemo should suffice for most applications
Spreading elements without changing the pending queue
(Nathan Forsdyke, Friday, August 02, 2002 12:46 PM)
The main disadvantage of Learning : Spread in subset operations is that all pending elements lose their pending status
You can open your element set in the browser and choose Child : Memorized before you use Learning : Spread. This way, only memorized elements will be processed
You can determine the length of the interval manually
(Aaron Koller, USA Educational, Friday, September 27, 2002 7:11 AM)
Is it possible for a user to manually determine the length of the first interval?
Use Learning : Reschedule (press Ctrl+J alternatively) from the element menu
You can transfer your learning material between
SuperMemo for Windows and SuperMemo for Pocket PC
(Marco van Damme, Netherlands, Nov 28, 2002)
I would like to export my learning material from SuperMemo for Windows to SuperMemo for Pocket PC
See: Revolution in the Pocket
You can easily inspect the number of outstanding
(C.Vermeer, Netherlands, Monday, November 18, 2002 9:40 PM)
I am a user of SuperMemo for Palm Pilot, who has recently bought SuperMemo for Windows. While learning in the Palm version, I have got used to the solution that each time I run it, in the first opening screen, the program shows if I have any outstanding material to rehearse. However, I do not see similar solution in SuperMemo for Windows. Is there an option that would make it possible for me to easily inspect if I have any repetitions to make on a given day?
You can inspect this information by peeking at the status bar, which is displayed at the bottom of the screen. If you do not see the status bar, there are two possible causes to that:
You can use SuperMemo for fast review too
(pirbos, Sweden, Tuesday, December 03, 2002 11:09 AM)
I would like to use SuperMemo for concentrated learning as well as long-term learning. I understand that relying on the SuperMemo algorithm for spaced repetitions guarantees my success in the long-term but is there a way to make repetitions before the date scheduled by SuperMemo (e.g. before an exam)?
SuperMemo 2002 introduced the concept of mid-interval repetitions, which makes it possible to execute repetitions ahead of time without interference of the review with the optimization algorithm used by SuperMemo. For example, if you would like to make repetitions on your Economics branch, which contains material critical for incoming exam, do what follows:
See: Subset learning for details
You can randomize repetitions
(Ole F., Germany, Monday, August 12, 2002 6:55 PM)
Can I randomize the outstanding items? I don't want to be shown the items in the same order that I entered them
SuperMemo 2006 and later will automatically add a degree of randomization to your outstanding queue. For this you must use Learn : Sorting : Auto-sort repetitions. You can set the degree of randomization Learn : Sorting : Sorting criteria. Remember to balance it with the priority of your elements. If you invest a lot of time in prioritizing your items, do not add excess randomness as it will ignore your priority queue.
In older SuperMemos, you can randomize repetitions with Learn : Random : Randomize repetitions (Ctrl+Shift+F11). Using Randomize repetitions is recommended if you experience breaks in learning or if you memorize more than you can review (e.g. with a frequent use of Postpone). However, you should never worry about the sequence of repetitions in moderate learning. Even for smaller bodies of knowledge, repetitions will get reshuffled due to different difficulty, memory strength as well as the noise factor introduced in the scheduling algorithm
Use Forget to skip pending elements that are of lower
(marcos, Tuesday, April 02, 2002 3:12 PM)
How can I shift less important pending elements to the end of the pending queue while learning with Advanced English?
When you see the element, choose Learning : Forget from the element menu. If you would like to put the element in the middle of the queue, type in the percentage (e.g. 50%). If you are not at the New Material stage, use Learn : Selected stages : 2. New material to resume new material repetitions
You can mix your top priority items, top priority
topics and other elements in 1:1:1 proportion
(Tomas Burk, Slovenia, Jun 24, 2005)
I understand that with the Intersperse function I can optimize the proportions of repetitions coming from different subsets of elements. However, this function is unclear to me. Could you explain it on an example? I would like to mix three subsets: top priority topics, top priority items, and randomly mixed " average" elements. How can I evenly schedule these subsets in 1:1:1 proportions in my repetitions (i.e. even proportions of elements taken from each set in turn)
To do this, follow these steps:
Once you master the above procedure and choose optimum shortcuts and icons, you can execute thus designed sorting procedure daily before repetitions in 30-40 seconds (depending on your fluency, the size of the subsets, and the speed of your computer).
Note that the above procedure assumes that your subset of top-priority knowledge is relatively small. For a very precise 1:1:1 mix you might begin with subtracting top priority subset from the outstanding subset. Otherwise, top priority elements will be slightly over-represented in proportion to their representation amongst outstanding elements overall.
Note also that Intersperse is random in nature. It will not let you do 1:1:1 with elements in a uniform sequence A, B, C, A, B, C, A, etc. Your sequence will be randomized. For example: B, C, A, A, B, B, C, A, C, C, B, etc. However, the probabilities of the occurrence of A, B, and C will be equal.
Retention vs. Consolidation in Workload
(Mike Butler, Jul 06, 2005, 01:09:09)
In the Workload window, under the heading of Retention, it will tell you your percentage of each day's retention; however, the next day another figure will be added somehow related to the final drill, i.e. 91.5 -> 76. What does the second number mean?
The first number refers to the retention of items repeated on that particular day. The second number tells you the "future" retention of these same items in future repetitions. As "consolidation" cannot be measured until items are repeated again (often years later), the number shows up as soon as the item with the shortest interval is repeated (if that item fails, consolidation stands at 0%, if it succeeds, it stands at 100%). The figure is updated each time an item from that given day's set is repeated in the future.
Retention tells you how well you are able to recall items today ("memory read"). Consolidation tells you how effective you were at reconsolidating items in memory ("memory write"). You get retention instantly. For consolidation value you may need to wait for months or years.
One of the most valuable applications of retention and consolidation data is in conjunction with sleep data where you can see how sleep deprivation affects recall (measured by retention) and consolidation (measured by back-retention). See: Poor Sleep = Poor Learning
Repetitions executed early are less effective
(malenfant, Thursday, August 25, 2005 3:39 PM)
If a given repetition has a long spacing, say 9 months, and the person does a review early, say at 3 months, what does that do to the schedule?
The earlier the repetition, the more pronounced the spacing effect. In other words, early repetitions are less effective in building memory stability. Hence the need to find a compromise between long intervals (forgetting) and short intervals (spacing effect). You can execute "forced repetitions" in SuperMemo 2004 and later and the repetition spacing algorithm will take the premature timing in the account. However, premature repetitions should be avoided if possible. If you are interested in a more mathematical answer, recently we have worked out an approximated formula that will help you estimate the consolidation effect of any repetition at any time with any past repetition history
You can randomize final drill with Tools : Randomize
(Wolfgang Flury, Tuesday, September 06, 2005 5:11 PM)
I've noticed that the final drill seems to get randomized only if I get there from a prior Learning stage but not when I directly select Learn : Selected stages : 3. Final Drill from the main menu. I was wondering if that's intended behavior
Yes. Final drill gets randomized only if the following three conditions are true:
For its lower priority, Final Drill should not rather be evoked separately from mainstream repetitions. However, if you wish to do it and have the drill randomized, use Tools : Randomize : Drill
After long break in learning, some intervals may seem
(Costanzo Cosentino, Friday, August 13, 2004 10:42 AM)
I have not used Advanced English for a long period. So I used the Mercy option. The problem is that looking at the next repetition date, I see that the next repetition is planned in 2007-2010 period and that is not good
If you return to repetitions after a long break, and you still remember your items, SuperMemo will consider those items as very easy and very well-remembered. It will then schedule the next repetition far into the future. This is a correct behavior and your chances of remembering those items in the future are as good as the setting of your forgetting index. However, if this worries you or some items are too important to be forgotten, you can use Ctrl+J to modify the next interval. After executing the repetition and noticing a very long interval, press Ctrl+J and provide a shorter interval
Starting "all over again" is possible, but not
(Stanley Ross, Sep 14, 2004, 14:54:53)
Is there a way I can reset learning to start all over. Sometimes I cannot study for weeks and I have a huge amount in the outstanding. Besides Mercy, can I just restart the learning process?
Yes. You can execute File : Tools : Reset collection; however, it is recommended you rather use Postpone to dilute the learning process and then Mercy to redistribute the outstanding material in manageable portions. SuperMemo is currently reasonably resistant to such brutal interventions in the learning process, and will still be able to help you make use of the previous learning; however irregular it was. If you reset your collection, not only will you entirely lose your former investment, you will also confuse SuperMemo each time you start learning seemingly "new material" that you have already been exposed to in the past. Here, SuperMemo is entirely defenseless as the reset collection will store no information about your past learning efforts
In SuperMemo, Memorized<>Remembered
(jj, UK, Sunday, December 24, 2000 1:54 AM)
I have noticed in the Statistics that the number of elements memorized increases even when I enter Fail when answering incorrectly. For instance, in the collection of US States Capitals, it was showing 100% memorized when I was still getting many of them wrong
Parameter Memorized indicates the number of elements in the learning process; not the number of elements you are able to recall correctly. If you make regular repetitions in the long run (i.e. over weeks and months), the number of elements you will be able to recall will approximately equal Memorized*Retention
Retention statistic assumes regular repetitions and
well-structured learning material
(dansujp, Sun, Sep 16, 2001 3:07 PM)
When I returned from vacation, I expected the retention to be something like 80% because I have not done any repetitions for two weeks. But it was exactly the same as before I left
The Retention statistic is derived directly from the measured forgetting index on the assumption of a negatively exponential forgetting curve. This curve is only representative of well-structured learning material. In addition, the forgetting index measurements are averaged over all recorded cases. A break in repetitions will invalidate the statistic. Resuming repetitions is not a guarantee of accuracy as the large number of earlier repetitions will result in overestimating the retention on a small-sample measurement. The only valid estimation of retention after a break in learning is the one that follows resetting the past forgetting index record (Tools : Statistics : Reset parameters : Forgetting index record). This will result in gathering new data that will approach true retention for the sample tested with accuracy proportional to the number of repetitions done
You can check the item's "creation" date in
repetition history (#2014)
(Sandra Jones, Sep 23, 2004, 16:30:55)
I'd like to suggest a right click date stamp in editing mode so I know when I added an item that might be outdated in the future
You can check that with Repetition History (e.g. Ctrl+Shift+H)
can learn "both sides" of a flashcard
(Cosmo, Germany, Feb 14, 1999)
Is it possible to learn vocabulary in both directions with one element? When I specify the question box as How are you? and the answer box as Cýýýýýýest can SuperMemo also change the directions of questioning so that I can train my vocabulary in both directions? Or do I have to define two elements?
Each question-answer pair is handled independently in the learning process. For this reason, you must define two elements. Usually the active element is repeated more often than the passive element but there are no fixed rules here. You leave it up to SuperMemo. You can quickly produce a reverse duplicate by using Duplicate (Alt+D) and Swap Q&A (Ctrl+Shift+S). The former will duplicate the element and the latter will swap the question with the answer (if you have many elements in the pending queue, you will probably want to follow it with Remember or Ctrl+M)
SuperMemo will not change the date until you close
the current collection (#28668)
(Chris Capel, Oct 03, 2004, 15:58:51)
If you leave your computer on through the night, SuperMemo's internal date will get stale. Is it important to restart SuperMemo in the morning?
SuperMemo will not change the date while the collection is opened. This is to prevent an unpleasant surprise of a sudden jump in the load of the outstanding material. The date will only be changed if you open another collection or if you quit SuperMemo. By default, the date change occurs at midnight. If you want to work after midnight, increase the setting of Midnight clock shift (hours) in Options. If you do not close your collections for a longer time, SuperMemo will display a warning
Your total working time will usually be far more than
your total recall time (#29029)
(Robyn, Oct 13, 2004, 10:12:16)
I am confused on your advice that very few individuals keep up more than 400 repetitions for more than a month. In Tools : Statistics : Simulation, 400 reps only takes 23.3 minutes per day, surely not so demanding? Also, you say that 100 reps would likely absorb at least 2 hours? But the total time in simulation is only 5.8 minutes?
You need to differentiate between the recall time (measured from the moment the question is displayed to the moment you press Show Answer) and your actual working time. Your recall time is likely to fall on average between 3 and 10 seconds per item depending on the type of the learning material. Your total working time will usually be far more. Apart from answering questions, you will browse, edit, delete, compare, search, add, etc. If you use incremental reading, the difference will be yet greater. It may happen that your average of 30 daily items will cost you 3 minutes in recall time and 2 hours of the total incremental reading time (most of it being reading and sifting through the new material)
Some Workload options are faster than others
(Marcos Pereira Rufino, Sep 20, 2004, 02:23:23)
When I finish my repetitions, I usually call Workload and double click the day to see all the elements I've just studied. I do this also with New items to revise new items I've committed to memory. The problem is that SuperMemo takes too long to display these items
If you choose Workload or Items or Topics in Workload options and double-click a day, SuperMemo will refer to dedicated files that duplicate database data for fast access. However, if you choose Repetitions or New items, SuperMemo needs to go through all elements in the collection to check if they meet the display criteria (by checking their last repetition date). SuperMemo might keep dedicated files that would keep the list of items repeated on a given day, but such files result in the swell of the collection size. In other words, some options in Workload will work faster, some will work slower (depending on the way data is accessed). The fastest way to review your today's progress is via the history browser (History on the element toolbar). History is also kept in a separate file. This is why it opens instantaneously
Qur'anic memorization and review
(Ayesha Nicole, Jun 01, 2011, 03:52:18)
Does SuperMemo have a reminder built-in as a pop-up message on the computer?
No. The assumption is that you learn with SuperMemo everyday. You need to remember to start the program. Once you start it, it will decide what material to show to you next to make sure you remember well.
Question and answer format is very effective
(Anthony P., U.K., May 2, 2013)
I stopped using SuperMemo because I read that question-and-answer format is not good for effective memory and intelligence. I started using clustering and concept maps, however, I have never mastered that art well enough to get back my high retention. Can you comment on this?
You fell victim of one of the prime anti-SuperMemo myths. It is true that you should cluster and structure information effectively in your memory. Mind maps might be useful in that process. However, to retain that structure in long-term memory, you need to make an effective review of its constituents. This is where you cannot go around SuperMemo unless you use the information regularly enough. In short: '''keep memories structured, review them as small bits'''. Question and answer format has proven very efficient and cloze deletions are currently the king of efficient review that combines speed of formulating with good retention. Come back to SuperMemo and focus on formulating your knowledge well