Pop-up dictionaries - how perilous are they?

[imron]

Personally, I think using a pop up dictionary to verify that the word means what you think it means is OK

I think this is danger territory and has a detrimental effect on long-term learning.  The reason being is that you probably half know the word and when you look it up there's a tendency to confirm to yourself oh, yes that's what I thought, and then move on (no need to learn this word, I already know it), and so you never address the fact that actually you don't know it with enough confidence to use it in a real world situation.

 

These almost known words are the ones you should absolutely be noting down and drilling separately.

[hedwards]

Personally, I think using a pop up dictionary to verify that the word means what you think it means is OK

I think this is danger territory and has a detrimental effect on long-term learning.  The reason being is that you probably half know the word and when you look it up there's a tendency to confirm to yourself oh, yes that's what I thought, and then move on (no need to learn this word, I already know it), and so you never address the fact that actually you don't know it with enough confidence to use it in a real world situation.

 

These almost known words are the ones you should absolutely be noting down and drilling separately.

 

 

I don't see a problem with that. Looking the words up in context is definitely preferable to drilling separately as you rarely get quizzed on individual words. What's more, because of the sheer number of words that have different meanings depending upon context, trying to drill separately is hugely inefficient without a lot of hackish behaviors.

 

I can definitely see your point there, but this criticism applies just as much to drills as it does to reference or even asking people for the word. Ultimately, as long as you form a guess before you look at the answer it will start to sink in. And ideally take a moment to reconcile the answer with your guess, it's not a problem.  Something I read recently suggested that a typical person has only about 28% recall on things that they're drilling You can boost that to more like 90% with mnemonics, but ultimately, drilling is frequently a poor choice.

 

Like I said earlier, the primary factor here is personal discipline, you can use these tools in a way that harms your progress or you can use them in a way that propels you along, but it's not the tool that's responsible, it's the lazy people that use them to avoid learning anything that are to blame.

[ouyangjun]

 

Quote

Personally, I think using a pop up dictionary to verify that the word means what you think it means is OK

I think this is danger territory and has a detrimental effect on long-term learning. 

 

I'm agreeing with imron on this one.  I used to use a popup that was C-E and would use it to confirm meaning...  I think this held me back.  After switching to the C-C I feel I'm learning better.

[querido]

To hedwards:

A program like Anki can practically force just about anyone to retain 85%.

Some tools like Imron's can include the whole sentence in which the word appeared, and this should be importable into Anki.  

[Guest realmayo]

The reason being is that you probably half know the word and when you look it up there's a tendency to confirm to yourself oh, yes that's what I thought, and then move on (no need to learn this word, I already know it), and so you never address the fact that actually you don't know it with enough confidence to use it in a real world situation.

 

If by use it in a real world situation you mean be able to produce the word in speaking or writing, then I agree, but I think most people have a much larger bank of words they can recognise than those they can produce.

 

If you don't use a pop-up, then lots of the time you won't look up that word at all. So you actually increase the amount of uncertainty in your head about it, and maybe trust your judgement even less the next time you see it. In a few instances, you sacrifice your reading flow in order to faff around looking it up, for only a small "win", i.e. confirmation that you were right.

 

On the other hand a quick pop-up check makes you think: "good, I am getting better with that word, and here's another context I've seen it in to add to my already-existing understanding of its usage" ... and move on. If it's a word I'm genuinely unsure of, rather than just double-checking, I'll note it (one click) and later see if it's in my flashcards: if it's not I'll add it, if it is, I'll reset it.

 

Again, I think it comes down to why you are reading. If it's simply to add new vocab to your brain and test your existing knowledge, then I agree, pop-ups aren't helpful.

 

 

Here's a confession: last night after reading a few chapters of a less easy (for me) author, I started to feel tired and wanted to go to bed, I couldn't be bothered to think hard about Chinese, but I also wanted to finish the chapter to find out what happened. So for the last couple of minutes of reading I just went pop-pop-pop-up though the difficult bits to get to the end. Is this so bad? Those last couple of minutes gave zero improvement to my Chinese. But I found out what happened at the end of the chapter. Without pop-up, I'd have just gone to bed. If you're doing that all the time, fine, it's a problem. But I'm not.

 

 

The irony with this discussion is that it's Imron's really fantastic software that is helping me improve the reading/vocab balance (by letting me take out and learn an appropriate number of words before I start reading).

[imron]

Something I read recently suggested that a typical person has only about 28% recall on things that they're drilling You can boost that to more like 90% with mnemonics, but ultimately, drilling is frequently a poor choice.

 

The problem I have with this sort of statistic is that usually it's referring to remembering random things - order of a pack of cards, numbers etc.  It's a completely different problem set from learning a language, and while yes, these techniques can also be applied to language learning, language already has a significant amount of hooks, connections and interrelations between words and the correct sort of drilling can be invaluable to improving your language abilities.  I used to be very dismissive of drilling until I tried it and noticed it made a huge impact.

 

Anyway, it's perfectly possible to drill with context - even without doing any sort of sentence extraction.  See this post for a more detailed description of the process I use, but basically all new vocab comes from context, and I relate everything back to that when drilling it.

 

 

Ultimately, as long as you form a guess before you look at the answer it will start to sink in

 

I tend to disagree with this, because for me at least that never really happened in practice.   Rather, I found the most effective way is once you've looked it up, you need to ask yourself why you needed to look it up, and then spend a few moments focusing on that and setting a trigger in your brain not to stumble on that same reason the next time you encounter that word.  If you're not looking at the reason why you failed to recognise a word and addressing that specifically, then it will continue to remain a stumbling block.

 

 

Like I said earlier, the primary factor here is personal discipline, you can use these tools in a way that harms your progress or you can use them in a way that propels you along, but it's not the tool that's responsible, it's the lazy people that use them to avoid learning anything that are to blame.

 

I agree that personal discipline is the primary factor, however tools that encourage lax discipline also have some responsibility.

 

 

The irony with this discussion is that it's Imron's really fantastic software that is helping me improve the reading/vocab balance (by letting me take out and learn an appropriate number of words before I start reading).

 

Chinese Text Analyser is basically my above arguments distilled in to software form It's designed to separate reading from vocabulary lookup, while providing tools to both pre-learn vocabulary before reading a text and also help you learn words you didn't know that you encountered while reading.

 

If you find that it is helping you achieve a balance between reading and vocab learning, then the software is meeting one of its major design goals, and it pleases me to hear that.

[Demonic_Duck]

I think this is danger territory and has a detrimental effect on long-term learning.  The reason being is that you probably half know the word and when you look it up there's a tendency to confirm to yourself oh, yes that's what I thought, and then move on (no need to learn this word, I already know it), and so you never address the fact that actually you don't know it with enough confidence to use it in a real world situation.

 

These almost known words are the ones you should absolutely be noting down and drilling separately.

 

I think this depends on your priorities. If you prioritise accurate active recall in writing and speaking, then of course what you say is correct. However, if you prioritise being able to read and understand books/newspapers/magazines etc., knowing the meaning (and ideally pronunciation) of a word in context is often enough. A quick glance at a dictionary definition may well be enough to cement it in place until the next time you come across it, with each subsequent encounter from then on strengthening your understanding until the point where you may well be able to use it just as well as if you'd drilled it.

 

Equally, you may prioritise different things for different words. Let's say you're reading an article about health issues: it may be important for you to be able to actively recall 哮喘 but not 湿疹, as you suffer from the former but not the latter. You'd still like to be able to recognise the latter in context, but you're unlikely to need to bring it up in conversation. As a savvy learner with only a finite amount of free time, you decide to add a flashcard for 哮喘 but not for 湿疹.

[imron]

If you prioritise accurate active recall in writing and speaking, then of course what you say is correct.

 

No, when I say real world situation, I am referring here explicitly to the real world situation of understanding it when reading it in some piece of native content.  Something which by definition you have failed to do if you had to look it up.

 

Meaning and possibly pronunciation are almost all you need to understand a word when reading, and while a quick glance may cement that in place, in my experience, it is unlikely because quick glances provide relief for the symptom (you understand what you are reading), but don't address the cause (why you didn't understand it).  That is, knowing the meaning of the word is not as useful as thinking and reflection about why you failed to understand it.

 

What tends to happen then is that you develop a reliance on looking up that word each time you encounter it, because you are always not quite sure enough - up until such a time as you decide to learn it properly.  The exception to this is when the looked up word is occurring with very high frequency just after you looked it up - e.g. several times in the same reading session.  If it's a word that might only appear once every few days, weeks or months then you'll more than likely be stuck in a perpetual state of looking it up.

 

This is why drilling can be so useful, because it helps cement these 'once every few days/weeks/months' words in to your memory long enough until you encounter them again.

 

Besides meaning and pronunciation though, there's one more important thing you need when reading - confidence that you know the word.

 

For example:

 

You are reading a book and you come across a word you can't quite remember.  You think about it for a few seconds and then you remember it.  You're not 100% sure though and there's a niggling feeling that maybe it's not right so you decide to check a dictionary (popup or other) and find that you were right.

 

What do you do next?  If you think 'yay I was right' and move on, you will more than likely doom yourself to always looking up this word, because you've failed to address the confidence factor.  If you stop and think 'hmm, why did I need to look this up if I was correct?  Next time I come across this word I going to remember that I remembered it correctly and therefore won't need to look it up in the dictionary', then you've got a much better chance of committing it to long term memory and not needing to look it up in the future.

 

Anyway, yes there are exceptions to this, and yes if you are being diligent you can avoid these sorts of problems, the problem though is the default mode for these tools discourage long term remembering.

 

It's like having an address book in your phone.  Back before mobile phones, I knew all the phone numbers of my friends and could dial them without any problem off the top of my head.  Nowadays the only number I know is my own.  I've delegated the task of remembering these things to my phone, and my brain has decided it has better use for the space than trying to remember those numbers and so doesn't put any effort in to keeping them in memory.  If you are regularly delegating your knowledge of words to a popup dictionary, and not doing anything else to supplement that knowledge, then your brain will do the same thing with the words you are learning - decide that it has better use for the space because you can always just quickly look it up again so there's not need to expend effort remembering it.

[hedwards]

I think this is all interesting debate, but I think there's a bit that we're ignoring. Some people just learn better by induction and others learn better by deduction. It's essentially the phonetics versus whole language argument.

 

I remember when I first started studying Chinese that I had high standards and I insisted upon holding myself to the hardest standard I could for the material. That was a huge mistake, it meant a ton of inefficiency because I was trying to avoid cheating myself. Unfortunately, it also greatly limited the amount of material that I would take in and it hobbled my ability to read as I'd try to translate the characters into sounds and then the sounds into meaning. And quite frankly in retrospect that was incredibly stupid. Even for more phonetic languages like English or German that's a poor idea, but with Chinese it's more or less fatal.

 

In my experience, it all comes down to finding something that works and sticking with it. Sometimes methods that seem insane turn out to be the best, even though logic suggests that they shouldn't work. Ultimately, as long as it helps me with input and or out put, I consider it fair game for use. And I've made a ton more progress for the effort now than I did back then when I was "doing it the right way."

[Guest realmayo]

Been interesting reading other people's thoughts. I do think there an obligatory place for 'book-only' reading, forcing yourself to spend time working out as much as you can before checking the occasional word in a dictionary. I'm sure that approach helps train the brain to work things out better, to read better. As Imron suggests, if you stop asking your brain to do something it stops being able to do it, and that includes figuring out hard stuff.

 

However, the sense I'd got previously from the forums that pop-up dictionaries are best avoided most of the time -- I'm still unpersuaded. Yes there are problems, but most or all of them can be mitigated (e.g. with good sense, self-control, adding looked-up words to a list that can be studied later, self-testing via flashcards).

 

I could happily see myself profiting from something like a trendy 80/20% split on time spent reading assisted/unassisted.

 

hedwards: what's the saying, 'don't let the best be the enemy of the good'? For me, I think that applies here. Because the amount of reading I'm getting done is, in the context of what I've been doing before, really good. 

[imron]

but most or all of them can be mitigated (e.g. with good sense, self-control, adding looked-up words to a list that can be studied later, self-testing via flashcards).

I can agree with this.  I'm not so much against pop-ups, but against using pop-ups without mitigation techniques.  I also think it's important to educate people not to just use popups blindly.

[Rufus]

Great discussion. This is some of the stuff I've been thinking about for a while. For instance, we are going to have some books available on Pleco where the reader can put it into the Pleco reader and tap on a word for a pop-up definition. Because of these reasons, I've questioned if we would be doing a disservice to learners who may abuse this "power" available at the touch of a finger. At the same time, there are lots of people who would use it legitimately when they need it. There is research on this area, but it's mainly for English language learning. 

 

Taking the entire discussion into account, what do you think should be the succinct recommendation for using pop-up dictionaries?

[imron]

I think Pleco reader is actually a great tool - because it's integrated with a dictionary and is almost no effort to add looked up words to lists for review.  It still requires diligence on behalf of the user to be clicking the 'add to flashcard' button.

 

My (not so) succinct recommendation for users of popup dictionaries is to be aware that just looking up definitions and moving on will probably not in itself help with learning words, and for the purposes of reading, looking up a word (even just to check) means you don't know that word well enough yet and it still requires further effort to learn (even if the checking showed you were correct).  To that end when using a popup dictionary:

 

Think about the word for a few seconds before looking it up.

Think about why you didn't know it after you looked it up.

Have an external system for keeping track of looked up words - especially words you were 'just checking' for later revision.

Make sure to also practice without a popup reader at regular intervals to assess actual ability.

If you are looking up too many words in a given session, consider looking for easier material.

 

The above don't need to be done for every looked up word, depending on comfort, interest and focus, but you should be making and effort to do it for a portion of all looked up words in a given reading session.

[mikelove]

Pleco-popup-wise, it's a fairly recent addition but in the current iOS/Android versions we actually have a 'reader' tab in the history screen; easy way to go back later and add unknown words to flashcards if you don't remember to do it at the time you look them up.

 

We've been looking at a number of ways that we might help people avoid lazy / thoughtless reader lookups, and one that will probably show up in our 3.2 update will be an option to have the popup bubble initially blank for words that are in your flashcard database (or a selected portion of it), with an extra tap required to reveal the definition. (still not sure what to do with Pinyin) We might introduce an option to highlight known words in some way in the future - I know those are popular - but I generally like the hidden definition approach better, since it injects itself into your reading flow only at the moment you're looking up a word and stays out of the way the rest of the time.

[querido]

I have a suggestion: Actually popup that flashcard, with the scoring buttons, etc., like a one-card flashcard session, retaining the score as usual.

The reader would have a chance to think "ah, I supposedly know this already", and a chance to get it right! 

[mikelove]

Heh, that's a fun one - maybe not as formal, but we could at least put handy little correct / incorrect buttons next to the definition after it's revealed in case you want to log it. Thanks.

[hedwards]

The more I think about it, this seems like a trade off between seeing a word many times and in many contexts versus trying to drill the same word and hope you get accustomed to seeing it in the right context. Personally, I tend to find flashcards where you have more than one or two similar definitions to be rather unwieldy, and yet a fair number of words have multiple seemingly unrelated translations and meanings.

 

For me, I'd much rather learn one or two meanings and then use pop up dictionaries for extensive reading for the other ones. That might just be me, but I tend to have a lot less luck with drills where I have to count on getting the word in a sufficient number of contexts to be useful.

[imron]

The more I think about it, this seems like a trade off between seeing a word many times and in many contexts versus trying to drill the same word and hope you get accustomed to seeing it in the right context.

The first important thing to realise is that you are only ever drilling words you have recently come across in context, so it's not really a tradeoff at all, it's not either or, you're doing both - seeing a word many times and in many contexts and also drilling those words to make sure you don't need to look them up next time you encounter them.  It's: come across a word in context, look it up, remember it in that context, add it to something for drilling, over a period of time drill that word and relate it back to the context, keep reading and probably encountering that word several times in context but without the need to look it up.

The second key thing to realise is that if you are doing regular reading, useful words will repeat in context regularly and this will complement any drilling you are doing.  Words that you don't encounter regularly are by definition not that relevant for you at that point in time and so it's ok if you don't learn them.  In my own experience, drilling of words I had come across in context markedly reduced the amount of the lookups I was doing when reading.

[hedwards]

OK, I think I see the disconnect. That's not really what drilling is for. Drilling is for verifying that the words you've already learned remain in your memory. They're also great for improving the fluency with which you use grammatical constructs you've learned. But, not for learning the structures in the first place. I suppose if I'm going to be thorough, there's not really any way around drilling when practicing pronunciation.

 

I know it's popular for people to use drills for learning vocab, but it's wrong to do that. I used to try that and it's the reason I failed miserably at learning my first several potential second languages. I'd recommend taking a look over at mnemotechnics.org because drills are a very inefficient way to pick up new words and they have a ton of information about how to memorize vocabularly efficiently.

[imron]

That's not really what drilling is for. Drilling is for verifying that the words you've already learned remain in your memory. They're also great for improving the fluency with which you use grammatical constructs you've learned.

This is exactly what I'm advocating.

 

I know it's popular for people to use drills for learning vocab, but it's wrong to do that.

For me context is the key.  Encounter the word in context, look it up, add it to a list for drilling to make sure it stays in my memory long enough that I don't need to look it up next time I come across it in the wild.