FLIC: another project to use technology to help in language acquisition

sounds pretty wild... hmmm, 'cuts language learning time by 50 per cent ... on 24-48 total hours courses'. Does this scale up though? '...only English, German, and Italian, but other languages can easily be added.' obviously the written aspect of Chinese need to be factored in First product to be launched end 2007, let's see what they come up with

For beginners it is just too hard if you are exposed to a variety of accents. I learned Mandarin in Sichuan where nobody speaks it. What a revelation when I went on a trip to Beijing and finally understood... More to the point, we had one teacher whose putonghua was not up to standard. That was just so confusing. The tones weren't always right, and then there was the s and sh and z an zh etc discrepancies. We started skipping her classes. There was just no point at that stage to learn from her, although the way she spoke was much closer to what we were exposed to on the streets. For us who live in in China I don't think it makes sense to have different accents in the podcasts. We get that anyway out in the streets.

I find the notion interesting to emulate the learning behaviour of a child. Unfortunately, the brain of a child is quite different to that of an adult. What strikes me is that those working with the EU project seem to be in need of language education themselves...

I remember reading somewhere about a theory that we all have a learning acquisition device (LAD) in our brains in early childhood. This disappears and then we are forced to spend the rest of our lives learning in a more structured manner, like adults. I think there is still debate about this, though. In my time at Goethe Institut some teachers seemed more keen on the "learn intuitively" approach while others felt structured drilling and explanation was the way to go. I learned more with the former than the latter, but this may have a lot to do with the fact that we all learn differently. Surely you've experienced in school and university that some teachers have a style that clicks with some students and not with others. Returning to the article about FLIC, I think they want to make learning more intuitive, rather than literally to make adults learn like children, but their writeup is a little vague.

Sounds like you may be refering to this? http://anonymouse.org/cgi-bin/anon-www.cgi/http://en.wikipedia.org/wiki/Generative_grammar

Oh no, not the generative crap again! They have no proof whatsoever undergirding their theories...

Sorry for the harshness of my wording, but I really feel strongly about this. I did look into it once with an open mind, but found no solid grounding, only suppositions.

I think Chinese and most other Aisian languages would take longer anyway, Sputnik, as they are really unfamiliar to Europeans. If you speak at least two European languages, a third (as long as you don't chose Basque, Hungarian, or Finnish) comes rather quickly. There is so much one recognizes, and the vocabulary necessary to get by is not so much. In fact, if you speak French and Spanish, Italian is perfectly intelligible. Learning a language in this context is more a formality to speak correctly, rather than a means to be able to communicate at all. One of the lessons I've taken away from studying Chinese is that it takes longer and more work to build up that "critical mass". I would imagine the same happens with Arabic or Japanese. When I'm satisfied with my Chinese, perhaps I'll tell you ;)

Did anyone catch what this project is actually doing? I'm confused reading their documentation as I don't understand what constitutes a "traditional" education in Chinese. They could be doing something quite interesting, or they could simply be measuring the difference between supervised and unsupervised learning. Or individual versus group learning. Off topic, but I've been disappointed for ages that there is funding available for this sort of thing, but almost none for the development of open texts / corpi which would benefit ALL language learners. As is, this project seems about as opaque as the NIST machine translation tests which don't actually release the competing translations for human scrutiny.

Some specifics buried within their documentation: Cross-cultural tests also revealed that in cultures where certain words are absent - for example the differences between 'L' and 'R' in Japan or 'W' and 'V' in the Indian subcontinent, small children can appreciate these differences, but once the child reaches eight months, the ability to distinguish has been lost. The FLIC method hopes to re-awaken these dormant abilities. Further tests on two and three year-olds show that incorrect grammar sounds simply 'wrong'. When asked to repeat sentences with correct and incorrect grammar, the children made more mistakes with the sentences containing incorrect grammar. To try and bring some of this childish wisdom back to adult populations, the company has devised a method of parallel processing in language learning, to give the new language an 'inner representation'. First, the student wears headphones, and repeats words. The student's voice comes through the left, and the trainer's voice through the right headphone. This allows the student to make direct comparisons. Secondly, the voices shift from ear-to-ear. This is important as it ensures both brain hemispheres attend to both voices - both the student's and the trainer's. This part of the CORDIS: News service http://cordis.europa.eu/fetch?CALLER=EN_NEWS&ACTION=D&... 2 von 2 10.05.2006 11:06 process is extremely important in making the language available to the subconscious mind - in much the same way as a child processes the information. Thirdly, the voices are broken up into four component frequencies, and blended, so each ear attends to two quarters of the student's own voice and two quarters of the trainer's voice. The voices, once blended, give the student a much enhanced appreciation of how his or her own voice should sound. Finally, the students undergo consonant enhancement. Certain phonemes are often difficult to pronounce. Phonemes are repeated, but with an 'incorrect' phoneme added. The incorrect phoneme is much easier to spot from this kind of direct comparison. The technology enables the consonants to be fine-tuned, by attending to the student's inner repetition. While the process seems to attend more to pronunciation, the evidence so far suggests that the complete learning experience is enhanced. 'We are using grammar without using grammar, hearing and applying words again and again,'

VARIETY - I think Cpod could certainly try some of these techniques in a podcast, for example on pronunciation. I think Cpod is also afraid (scaredy cats! nah nah na nah!) of presenting Mandarin in different accents. Although one of the techniques above was to feed the students own voice into a real-time comparison with proper pronunciation is hard to do in a podcast, would it be so hard to present side by side the way Canadians mangle a Chinese word, whereas how U.S.A'ns add a beautiful lilt to a word said by a native speaker? I do think such comparisons might hold some possibility for helping us more quickly lose some of our 'accents'.

Ok. This is interesting. I just reread the press release. It has a lot more detail on what they are doing than the documentation I saw on the website.

Direct feedback to improve learning FLIC uses a feedback system, equipping learners with a headset (earphones) and microphone, plus a black box (or programme, in the case of the home version), which mixes the sound. When learners begin a FLIC course, they start by reading text while listening to it being pronounced by a model voice in the earphones. Next, they mouth the text while listening to the words (“The brain is working although the voice is silent,” explains Warnke). Step three is actually speaking the words out loud. Then – and this is the clever part – the system plays back the learner’s efforts in one ear, and the model voice in the other ear. “And it doesn’t stay in the same ear, but moves from ear to ear, so that inter-hemispheric contact is enhanced,” explains Warnke. “Both hemispheres are very important for language learning. The left hemisphere is involved in word recognition, it’s the word processor of the brain; but the right hemisphere rules prosody – a vital decoding process that uncovers non-explicit shades of meaning.” Using the learner’s own voice aids learning, because research shows that we pay most attention to voices like our own. So, in the next stage, FLIC uses this fact by modifying the model voice, blending it with the learner’s voice, which has been divided into bands and analysed for speed, pitch, and so on using software developed by Stockholm’s KTH, one of the FLIC project’s partners. The resulting voice mimics the learner’s own, yet has the superior pronunciation of the model voice. “When learners hear this, they experience themselves speaking a foreign language much better than they expected,” says Warnke. “They are then happy to keep practising and speaking.” Helping build and maintain vocabulary An innovative use of voice techniques is also used to improve vocabulary retention. Learners hear a word in the target language in one ear, and its meaning in their own language in the other ear, simultaneously. “Again, the words and translations switch between the ears, and so the hemispheres,” says Warnke. Learners listen and participate in dialogues, picking up the structure of the language by example, without ever having to endure a grammar lesson. “They absorb the rules naturally, which is exactly how small children learn languages,” says Warnke. “FLIC

I think Lantian brought up a good point concerning exposing learners to a variety of accents. In the ESL classroom, listening is the toughest component for students because a) English has many different accents and b) the way teachers speak in class is not the way native speakers speak outside of the classroom. Why? Because in the classroom ESL teachers must do what's called "grading their speech". This is essentially where the teacher eliminates all use of colloquial language, word fusion / connected speech, and speaks slowly and clearly so that students can understand. This is crucial for learners up to the Intermediate levels. What disadvantages does this have? Well, the students in a sense become spoiled by the teachers in-class language. They're effectively listening to unnatural natural speech (if that makes any sense). In short, when they actually run into a native speaker on the street who has not learned how to grade their speech or speak with a language learner, the native speaker generally speaks over the student's head. (For us Mandarin learners: compare Jenny's podcast accent with that random taxi driver's.) So, what are some advantages / disadvantages to CPod presenting a variety of accents? Perhaps the most obvious advantage is that learners would be exposed to a variety of not-so-perfect / standard accents, which is what we generally run into outside of the classroom / away from our headphones. A disadvantage, however, may be that learners begin to imitate the "wrong" accent, and therefore develop bad habits / a bad accent. What does everyone else think?

I'm not sure it's the same crap, or a new kind, Go ;) Sometimes it's hard to distinguish all of these theories, especially as some of them have several names. I remember an educational method which was used some places to teach children how to read. They would show whole words and read them out, rather than start by teaching individual letters. The idea was that sooner or later an "aha!" moment would occur and everything would fall into place for the student. Apparently that approach has fallen into disrepute since then... Still, let's keep an open mind. The Finnish education system is one of the most successful in the world, with first place in the OECD's rankings. One of it's distinguishing features is that students aren't taught anything until they are seven. Before then they are just allowed to play together. Apparently any loss due to studying structured content later is offset by cognitive gains resulting from early socialization among peers and free game play. May be counter-intuitive at first, but it makes sense when you think about it.