Seven language learning myths, and what the research says

Myth 1: You learn best in your own learning style

The idea is that some people are visual learners and others auditory or kinaesthetic, and that teaching to your style makes you learn faster. In 2008 four psychologists, Pashler, McDaniel, Rohrer and Bjork, reviewed the evidence for the Psychological Science in the Public Interest journal. They set out what a fair test would need: people with different styles, taught with different methods, and an interaction showing each group doing best with its matching method.

They found virtually no studies that showed that pattern, and the few that were designed properly contradicted it. Their conclusion was that there is no adequate evidence base for using learning-style assessments in general education. They added that many versions of the idea have never been tested at all.

What is true: people have preferences, and enjoying a method helps you stick with it. Preference just isn't evidence that a method teaches you more.

Myth 2: You can learn a new language in your sleep

Sleep does help memory, which is where the myth starts. The studies are narrower than the headlines.

The practical version is modest: study before bed, then sleep. Leaving a podcast on all night will not teach you a language.

Myth 3: The FSI hours tell you how long your language will take

The US Foreign Service Institute's difficulty list is the most shared chart in language learning. On the State Department's own page, the times are the hours usually expected for its students to reach a score of 3 in speaking and listening on the Interagency Language Roundtable scale, a professional working level. A typical FSI week is 23 hours in class plus 17 hours of self-study.

The page itself warns that the time varies with a learner's natural ability, earlier language-learning experience and time in the classroom. So the chart is a fair guide to relative difficulty for English speakers: Korean really does take longer than Swedish. It is not a forecast for someone studying 30 minutes a day, and it says nothing about reaching an earlier goal such as reading graded stories at B1.

Myth 4: Raising children with two languages confuses them

Parents are still sometimes told to drop a language so a child doesn't fall behind. Krista Byers-Heinlein and Casey Lew-Williams reviewed the research in 2013 and found the opposite. Bilingual infants tell their languages apart and show no sign of confusion. Mixing two languages in one sentence is a normal part of bilingual development, not a symptom.

Bilingual children are no more likely than monolingual children to have language difficulties or delays, and when you count the words they know across both languages, their vocabularies are about the same size. The authors point out that testing a bilingual child in only one language makes a false delay more likely, and that advice to stop exposure to one language isn't supported by the science. Our page on reading as a family has more on children and second languages.

Myth 5: Using your own language is cheating

Many courses still work on the assumption that a new language is best taught and learned only in that language. In a 2012 review for Language Teaching, Graham Hall and Guy Cook describe that monolingual assumption and note that it has been increasingly questioned.

One of the experiments behind the change: Batia Laufer and Nany Girsai taught 10 words and 10 collocations to three groups of high-school learners for the same time. One group used the words in meaning-focused tasks, one did form-focused exercises without translation, and one contrasted the words with their first language and translated. The translation group did significantly better on every test, immediately and a week later.

That doesn't make translation the whole method. It means a quick look at your own language, a gloss or a tap on a word, is a tool rather than a failure. The trade-offs are on our page about parallel texts.

Myth 6: Babies pick up a language from a screen

Patricia Kuhl and colleagues tested this directly in 2003. American nine-month-olds from English-speaking homes had 12 lab sessions with native Mandarin speakers. Afterwards the babies could distinguish a Mandarin sound contrast that babies of that age normally stop hearing.

Other babies got the same speakers and the same material by video or audio recording. That exposure had no effect: they did no better than babies who heard only English. The authors concluded that infants learn speech sounds from live but not from prerecorded exposure. The study was about speech sounds in infants, not about older children or reading, but it is a clear warning against leaving the teaching to a screen.

Myth 7: You have to learn the grammar before you can read

In a study published in 1983, Warwick Elley and Francis Mangubhai reported on a "book flood" in Fijian primary schools, which gave classes a supply of high-interest English storybooks. Pupils who read them progressed in English reading and listening comprehension at about twice the rate of pupils without them. Reviewing book-based programmes in 1991, Elley concluded that children immersed in meaningful text, without tight control over syntax and vocabulary, appear to learn the language incidentally.

Reading at the right level also hands you grammar before anyone explains it. This A1 sentence puts the verb second, after the time word, which is a rule of German word order that you meet long before you study it:

Grammar study still helps, especially for accuracy; our page on learning grammar by reading weighs the evidence. The myth is the order, not the grammar.