Cognitive Daily

Toddlers learn new words at an astonishing rate—an average, according to Steven Pinker, of over a word every two hours. Yet attempts to drill children to improve vocabulary are often frustrating. Kids seem to learn words better through observing the environment than they do by rote. So what exactly are they observing?

One possibility is that the child is paying attention to what others are looking at: if a grown-up looks at a construction site and says “look at the bulldozer,” maybe kids learn “bulldozer” because they have learned to follow the grown-up’s gaze. Another possibility is that kids assume that the object they don’t have a word for is the one being referred to: if the child sees a tree, a bird, and a big noisy yellow thing, then when the adult says “look at the bulldozer,” the child assumes “bulldozer” means “big noisy yellow thing” because she already knows that a tree is a tall green thing and a bird is a fluffy flying thing that eats worms.

One of the symptoms of autism is that people with autism don’t perceive the intentions of others—so if we learn language only by observing the gaze of others, it would seem that kids with autism would have more difficulty learning language. Mellissa Allen Preissler and Susan Carey developed an experiment to see if toddlers with autism could use the speaker’s gaze to learn the names of objects. They gave 24-month-olds (some with autism, some normal) objects they’d never seen before (a doorstop, soap dish, tire gauge, or cheese grater). The experimenter held a different new object, and gazed at that object while the toddler was playing with his or her object. Then the experimenter named the object in her hand with a nonsense word (“peri”). She placed her object and the toddler’s object in a bag with two other objects. Then she asked the toddler to “find the peri.” Children with autism most frequently chose the object they themselves had been holding, but normal children correctly identified the object the experimenter had been looking at 70 percent of the time.

In a second experiment, kids were presented with two objects—one familiar, and one unfamiliar (as reported by the children’s parents). The experimenter asked kids to “show me a blicket,” using a nonsense word to refer to the unfamiliar object. In this case, both normal children and children with autism were able to successfully identify the novel object. Here’s a summary of the results of both experiments:


So though children with autism are often unable to follow the gaze of a speaker to learn what object he is referring to, they are just as able as normal children to use the process of elimination to associate a word with a new object. Though social cues such as observing the gaze of another are important in learning language, the same concepts can be learned other ways. Perhaps what’s most amazing about kids’ ability to learn language is how effortless it seems, even for kids with impairments at making inferences about others’ intentions.

Preissler, M.A., & Carey, S. (2005). The role of inferences about referential intent in word learning: Evidence from autism. Cognition, 97, B13-B23.


  1. learning

  2. #2 natalie
    May 8, 2006

    i think sarah brown-schmidt at illinois has been doing work on eye-gaze in english monolingual adults determining referents when hearing chinese.

New comments have been disabled.