杏吧原创

Computer model explains toddlers’ chatting

The way children rapidly learn a vocabulary may be a result of the way most languages are structured

YOUNG children become chatterboxes within months of speaking just the occasional word. Now one scientist thinks he knows why.

Parents of small children will be familiar with the so-called 鈥渨ord spurt鈥, when a child goes from barely talking to suddenly uttering hundreds of new words, sometimes after hearing them only once. have been suggested to explain this phenomenon. For instance, perhaps learning a few basic words helps a child learn others, or maybe children quickly understand that if there is a word they don鈥檛 recognise, it belongs to the object they can鈥檛 name.

Now language psychologist Bob McMurray at the University of Iowa in Iowa City has a simpler explanation. He believes that the acceleration in a child鈥檚 learning is down to the way most languages are structured. All languages, he says, contain a distribution of words where most are of medium difficulty and a few are either very easy or very difficult. Children learn all words in parallel, McMurray adds. He factored these rules into a computer model that simulated how long it would take to learn 10,000 new words.

Each simulation produced the same pattern: essentially, a child masters the small number of easy words, but has simultaneously been working on the medium difficulty words. Soon afterwards they master these as well, but it feels like a burst because there are more of them. 鈥淎cceleration is an unavoidable by-product of variation in difficulty,鈥 says McMurray ().