How Learning works

Why stories

You finish stories. You rarely finish exercise sets. Almost everything else on this page follows from that one difference.

Narrative sticks

Narrative text is remembered and understood better than expository text. The researchers behind that finding attach a caveat we will repeat: it is not a licence to force every subject into story form. Grammar tables still have their place, and you will find them here too.

Where Krashen sits

Plenty of language products rest on Stephen Krashen's input hypothesis, the idea that understandable input alone drives acquisition. That position is contested.The evidence pointing the other way is the instruction and feedback line. Focused, explicit instruction produces large gains, and corrective feedback holds up on delayed tests. Input plus targeted correction is the shape of a course here.

How the parts fit

A course starts from a story you would read anyway. Your weak rules go into its sentences. Exercises follow each chapter, and the words you miss end up in a review queue.Each piece has its own page below, including the parts where the evidence is thin and we say so.

Getting the most out of it

  • Pick a story premise you would read in your own language. Interest is the variable you control most cheaply.
  • Finish a chapter before opening its exercises.
  • Read the references index if a claim on any of these pages sounds too tidy.
  • Send us the claims you think are overstated. The gate that produced these pages exists because it is easy to overstate them.

Under the hood

Every claim on these pages went through three independent verification votes against its primary source. Claims that failed were dropped or rewritten, and the rewritten ones carry a fixed wording we are not free to improve on.Where a source is funded by a company, the references index labels it as such. Where a finding rests on a single small trial, the page says so in the sentence that uses it.

Sources

  • Mar, Li, Nguyen & Ta (2021) Memory and comprehension of narrative versus expository texts: a meta-analysis. Psychonomic Bulletin & Review.
  • Norris & Ortega (2000) Effectiveness of L2 instruction: a research synthesis and quantitative meta-analysis. Language Learning.
  • Li (2010) The effectiveness of corrective feedback in SLA: a meta-analysis. Language Learning.
Effect sizes, study designs and source labels
Browse Learning