GUEST POST
Andrew is a Cambridge CELTA-certified English as a Foreign Language Teacher with 20 years of experience specialising in advanced professional English for B2–C1+ learners, including corporate managers, founders, engineers and researchers working in multinational environments.
By Andrew P | Cambridge CELTA-certified English Teacher | Sherpa Tutor
Does the following sound familiar? You study a vocabulary list or do a couple of units in a grammar workbook, you think you know the content, but you find you can never produce the words or grammar structures you learned when you're in a real conversation?
This is a predictable feature of how memory works. Firstly, new information is understood in relation to known information, so studying content in isolation falls flat. Secondly, new information creates new neural connections, and these connections fade over time as the brain tries to eliminate untested ideas and relationships to keep you safe, so unless you review new information (preferably in new contexts) your study content will be lost forever.
In 1885, psychologist Hermann Ebbinghaus ran an experiment on his own memory. He memorised lists, then tested himself at intervals to see what information his memory retained. He found that we lose roughly 50% of new information within the first hour, then we lose about 70% within a day, and less than 20% remains after a month. This fall-off of information has since been described as the forgetting curve.
This means that you are not scaffolding new information on top of previous information and you aren't building rich neural networks of interconnected cues and associations. It's like starting from almost zero every time you study.
Most people quite naturally think of language learning as a skill they can iterate on over time. "I'll spend 2 hours Sunday afternoon and learn 60 new words, or do 3 units of English Grammar in Use," they think, but by Tuesday up to 80% of what they think they learned is gone. This massed practice has been shown time and time again, most notably in a 2006 meta-analysis by Cepeda and colleagues, to consistently underperform distributed practice, sometimes by margins of 200% or more.
The problem isn't effort. It's timing. Reviewing the same word five times in one sitting only trains short-term retrieval. It feels like learning because the word is still hanging around in working memory, but it isn't building anything permanent.
Fifteen minutes a day for thirty days will beat seven hours in one weekend, not because of the total time, but because of when that time is spent.
In the real world, life doesn't present you with multiple ways to say something. You have to produce the content words that match your ideas, insert them into the right grammatical framing, and make sure the tone is right, all in under a second, or you miss your opportunity to speak.
Production-focused practice, not casual chat, but targeted, time-pressured production of previously-learned information, produces significantly stronger long-term retention than recognition-based drilling. Students can often recognise thousands of words they simply cannot produce in speech.The only thing that resolves this is the method I developed: frame-based retrieval.
Take three simple words: ask, request, and demand. These aren't three synonyms to memorise separately. They're one frame with three settings, differentiated by formality, power dynamics, and social context, and crucially, each setting comes with its own grammar. "Request permission from" behaves differently to "ask about," and that syntax isn't arbitrary; it reflects the underlying meaning.
Well-timed exercises that connect words and grammar to the emotion of a communicative goal within a precise context are the key to developing advanced fluency, and this can only be done with a skilled teacher in a one-to-one setting where the pressure is real.
It sounds counterintuitive, but cognitive science has shown that making practice feel harder, forcing intuition-based responses due to the subjective nature of the exercise, actually makes language stick. This is because of the broader neural stimulation triggered. Repetition exercises build upon the neural trace left behind, pulling new language into the productive type, rather than the language you can read and understand but not spontaneously produce.
Three things I build into every review task:
In the first few weeks, students often say this doesn't feel like learning, because there's not always a "correct" answer. But once they start finally producing new language in the moment, with genuine mind-to-mouth fluency, they realise that they've been using the wrong approach to learning English since they graduated beginner level, and they never look back.
If you'd like help building genuine speaking fluency rather than just vocabulary recognition, I'd love to help. Get in touch through my profile today to find out more.
Andrew P
Tutor
English Fluency Lessons | Cambridge English Tutor of 15+ years
Sherpa has hundreds of qualified and experienced UK tutors who are ready to help you achieve your goals. Search through our tutors and arrange a free 20 minute introduction through our industry-leading online classroom.
Find a TutorSimilar Articles