Who here uses the Review Sentence function? How useful do you find it?
Do you prefer the matching pairs activity or the reorder-the-sentence activity? Why? Do you find them doing much for your language skills?
It looks like Duolingo ones.
In a real conversation, there are no pre-presented words, so I have doubts about the effectiveness of the activity, I rather listen and read more. My primary learning activity is composed of LingQ (Reading / Listening) and Anki. Anki helps me by exposing sparse words more often than just reading.
2 Likes
That in my opinion is LingQ attempting to emulate Duolingo, Busuu, Babbel and others.
I started using LingQ after completing the Babbel German ācourseā. I used those exercises for six months for German only. At that time I was scared of going it alone, I felt I needed guidance. I was also rather naive about language learning, despite more than ten years of language learning prior to that. Perhaps we all have to go through a phase where we learn how to learn. I feel that I spent between one and two years doing so, and ended up largely returning to the so-called traditional techniques Iād used thirty years earlier! The one big difference is that I now have access to huge amounts of input.
Those exercises are essentially an SRS, and my experience is that you cannot learn a language using SRS tools alone. I donāt doubt that SRS tools are a very good way to memorize facts, and they can be used alongside reading and listening to texts. But they should not be used on their own. Looking back, I would not have used them had I known what I know today.
I once asked Steven Kaufmann, in the comments section of one of his videos, why LingQ includes those tools given that they go against his own philosophy of language learning. He said that it was to support people who like using them.
4 Likes
I 100% agree. We can not learn effectively only with Flashcards, but itās a nice supplement.
1 Like
From his YouTube videos, Steve does mention often that he doesnāt use the SRS. However, he has mentioned several times he does use the Review Sentence function after reading the sentence in Sentence Mode though.
Iām particularly curious how people find these specific exercises of matching pairs and the reorder-the-sentence functionalities.
How odd, then I have no idea why he answered me as he did. Iāve only watched a few of his videos, maybe ten. I eventually got sick of internet polyglots and I stick mainly to linguistics text books.
I did use the matching pairs and reorder the sentence exercises. I didnāt find them very useful. For me there are two problems with them. The first is that they rely on short term memory i.e. you have been introduced to the sentence and then you are tested straight away while it is fresh in the short term memory. That doesnāt help it move into long term memory. That is expected by the research presented in the book āMake It Stickā. The second problem is that these exercises are abstract, it is quite possible to do them without thinking about the underlying meaning i.e. they are abstract associations.
When I come across a new word or word phrase, I prefer to focus on the meaning. So a Google on passerelle throws up pictures of foot bridges. If I Google Tu en as encore sous le pied I find explanations of the saying, and the origins. And I then play around with the words. It works for me. Spending time researching with Google is fun, and works far better than automated tools.
1 Like
The issue is that thereās a gap between what practitioners are doing and what researchers have looked into. Personally, I prefer to spend more time listening to practitioners with only dabbling small amounts in the research.
I agree. I read Make It Stick in the end, though I was quite hestitant to due to the publication date (2014). A lot of psychology was turned on its head in those and the coming years with famous studies failing replication. However, I think most of the stuff in the book still holds, with only a few things mentioned failing replication (such as the marshmallow effect).
I was playing around with going straight to the Sentence Review without the initial reading of the sentence. This would get over the fact that the words/sentence are still in working memory. It would also make you think about the underlying meaning of the words too (though you can still use cues of capital letters and punctuation to help you out though).
I find it not so bad when you have a bunch of new vocab to practice matching to reinforce ahead of practicing each sentence. But as the amount of new words go down you can skip to get to the sentence recording. Obviously if it is a list or really long I will skip the match / sentence reordering entirely just to do some dictation practice/shadowing. I agree though it is the most Duo-esque thing on here.
1 Like
And yet some months ago when I suggested to you that we could learn a lot by studying polyglots, you dismissed that idea.
Which polyglots do you listen to, and what do you learn from them?
For what itās worth, I listen to a few polyglots, such as Richard Simcott and Alex Rawlings. They both work as language teachers which greatly informs their commentaries. I donāt doubt the skills of Kaufmann, Olly Richards and others, but their videos are product marketing.
I find linguistics fascinating especially research that tackles the underlying processes of learning an L2, and an L1. You are right that there is a discord between research, and practical language learning. That said, Iām interested in finding out what we know with reasonable certainty. The field of SLA is a bit of a mess, as there are so many competing theories, with too little agreement.
As Iām sure you know, language practitioners and language learners donāt have to be mutually exclusive. Many linguistics researchers are accomplished language learners themselves. Perhaps it is implied in your remark, that a language researcher should ideally have advanced competence in at least two L2s, preferably quite distinct ones, so that their theories are grounded in reality, unlike those of Noam Chomsky and Stephen Krashen.
Yes the marshmallow experiment and the one with prisoners and guards have received considerable criticism. As you say, Make It Stick seems to be on solid ground as they cite numerous studies to back up their claims. Incidentally, the book was recommended by Steven Kaufmann in one of his videos. SK is well read.
Some months ago someone on a forum sent me a private message. That person is a native English speaker, and completely fluent in Dutch and French, and probably one or two other languages, such that they can be mistaken for native speakers in at least two L2s. He said that there is not really much to discuss about language learning, itās straight forward. You just need to find some way to get words and grammar into your brain, and then expose yourself to the language as much as possible. Each polyglot has their preferred method of doing that. Of course we can discuss tools, such as Anki, or our preferred way of listening, but consistency and hard work are key.
Incidentally thank you for the Anki tip. I find myself going through entries much more quickly, and focussing more time on the ones that donāt sink in. Iāve also come to the rather obvious realisation that most apps, such as Duolingo, are little more than glorified SRS, which is why they donāt work very well.
1 Like
I have watched nearly every YouTube video Steve Kaufmann has released in the last three years since joining LingQ (plus I watched some before this, etc.). These days they usually arenāt super interesting as itās the same old same old, but occasionally thereās a video which I get something out of it. For instance, knowing that Steve used to wait before hiring a tutor to practise talking until he reached ~10k LingQ words, but these days has dropped it down to 3k-5k or so is worth noting. Or how Steve thinks about language learning in the phases of "absolute beginner ā plateau ā take-off."
Alexander Arguelles also has some interesting things in his YouTube videos to say such as how he manages all the languages he knows or him showing how he learns a new language or him coming up with his sonās plan to revive Korean.
I also watched quite a few Polyglot Conference talks back in the day.
I think as an individual I can learn from other language learners and polyglots (to save reinventing the wheel, learn from the trials and errors of others, right?), but from what I thought we were discussing then is what research can learn from them, unless I remember incorrectly? Science and research can learn a bit, like preferences for methods and to what degree learning many languages changes the brain, etc., but for science to understand the fundamentals of how language learning works, they probably want to be looking at the more common people, who only know one or two languages.
You also need to focus on your accent too. Without that, no amount of fluidity will ever result in someone thinking you are a native speaker.
I think they are worse than a glorified SRS! 
DuoLingo may have changed since I last played with it several years ago, but you donāt practise stuff you learn from lessons before.
Iāve probably gone through the same process of watching as many videos as I could, then after a while they all sounded the same.
I was suggesting that both science and learners could benefit from studying polyglots. These days I am not convinced that we (learners) benefit much from their advice, beyond the recommendation to get a lot of input which I think we all agree on, as they usually promote a magic method and a product.
Yes it would be instructive to know if and how learning multiple L2s changes the brain. Does learning two or three L1s change the brain? Does learning an L2 change the brain? Are there changes in the brains of people who learnt three L2s that are not present in those who learnt one L2?
But what I was suggesting is that polyglots might be more efficient learners. Now I think we agree that there is no magic method, as they all use different techniques. However studies might be able to determine whether or not they are more efficient than non polyglots, though such studies would be hard to carry out, and of course they would require funds. And if they are more efficient, the next question is why. Perhaps changes in the brain are the reason, perhaps it is the methodologies, maybe both.
Iām pretty sure that were I to start a third L2, I would progress much more rapidly, mainly because Iād not use Duolingo, Babbel etc and Iād use the methods I have developed, which of course are nothing new or special, but they suit me.
Of course. Not much attention is paid to that aspect.
I spent yesterday walking in the countryside - four hours plus a brief lunch break ! - with a friend who is using Duolingo for French, and Iāve recommended that he gradually moves over to podcasts, initially ones for learners of course. Itāll be interesting to see his progress. He is 70 years old, but that should not be an impediment.