Naming of Objects and Object Parts

"The Clothespin"

How do we represent objects for the purposes of naming, and how do children learn to name objects?
Philosophers such as Quine have pointed out that the novice language learner faces a problem of "indeterminacy": Hearing some word in the context of some event, the child might interpret that word in an infinite number of ways, only one of which will be consistent with the intended meaning. How does the child language learner solve this problem?

Research over the past few decades indicates that toddlers come to the word learning task with biases to
interpret novel object names as names for members of categories such as animals, artifacts, etc. Perceptual
information provides surprisingly rich information that can help toddlers assign objects to categories, and
thereby help them generalize object names correctly (see Landau, Smith, and Jones, 1998, for review).
For example, consider the two panels below. Each shows an object on the left, then three changes in shape.
If the first object in each panel was named a "dax", would you generalize this name to any of the three objects on the right?

Does your answer differ for the two panels? (See Landau & Leyton, 1999 for discussion)

Now consider how you would decide how each of the objects shown can be divided into its main parts. Do we use
pre-determined components, or do we use properties of an object's contour (as suggested by Hoffman & Richards, 1984). How does such knowledge affect the naming of parts? Does the functional nature of an object affect how we name the parts? Current work in our lab seeks to explore such issues of the relationship between perceptual information and naming byexamining the role of perception in naming-- how perception might support naming, and how naming may go beyond perception. For a recent review, see Landau, 2001.