Audiobooks alone don’t teach vocabulary
MIT researchers found that audiobooks by themselves don’t significantly boost vocabularies for struggling readers, but pairing audiobooks with one‑on‑one instruction does help learning. That result underscores a craft principle for audio storytellers: sound is powerful, but retaining complex information often requires scaffolding or repeated, varied reinforcement. (news.mit.edu)
Kids can pick up new words from a story the same way they pick up a recipe by watching someone cook: they catch some steps, but they miss a lot if nobody stops to explain the tricky parts. An MIT study tested that idea with 314 third- and fourth-graders and found that listening alone was not enough for the weakest readers. (news.mit.edu, onlinelibrary.wiley.com) Vocabulary is the stockpile of words a child can recognize and use, and that stockpile grows when books keep introducing unfamiliar words in context. Children who struggle to decode print often miss those chances, so researchers wanted to know whether audiobooks could replace some of that lost exposure. (mcgovern.mit.edu, onlinelibrary.wiley.com) The team launched the project in 2020, when Covid-19 school closures made remote learning urgent and widened fears about the “summer slide” effect for poor readers. John Gabrieli, Ola Ozernov-Palchik, and Halie Olson built the study so families around the United States could participate from home. (news.mit.edu, mcgovern.mit.edu) The experiment lasted 8 weeks and split children into three groups. One group got text-supplemented audiobooks alone, one group got the same audiobooks plus one-on-one guided support, and one active control group did not get the audiobook intervention. (onlinelibrary.wiley.com) Text-supplemented audiobooks are recordings paired with the printed words, so a child can hear the sentence and see it at the same time. That setup gives access to grade-level stories even when sounding out every word on the page is still hard. (news.mit.edu, mcgovern.mit.edu) The one-on-one support acted like training wheels for word learning. MIT trained college-student tutors with no formal education background to pause on hard words, explain meanings, and use proven teaching methods during remote sessions. (mcgovern.mit.edu, (mcgovern.mit.edu)) Across the full sample, both audiobook groups improved on book-specific vocabulary, while the active control group did not. But the split that mattered most was reading ability: poor readers benefited only when the audiobooks came with that one-on-one scaffolding. (onlinelibrary.wiley.com, news.mit.edu) The scaffold group also spent more time listening to the recommended books during the study. That means the added human support did two jobs at once: it explained unfamiliar words and kept children engaged with the material for longer. (onlinelibrary.wiley.com) This fits a broader reading-versus-listening pattern researchers have seen before. A 46-study meta-analysis summarized in The Conversation found that comprehension is often similar across reading and listening, but reading pulls ahead when people can slow down, reread, and check details at their own pace. (theconversation.com) MIT’s paper was published in Developmental Science on March 17, 2026, and the result is narrower than “audiobooks work” or “audiobooks fail.” Audiobooks opened the door to new words, but the children furthest behind learned the most when another person helped them walk through it. (onlinelibrary.wiley.com, news.mit.edu)
