Psychology

A connectionist account of the object-substance distinction in early noun learning

Description
Abstract Young children learning vastly different languages generalize nouns in the same systematic way. Young children extend the name of a solid object to other objects that match the exemplar in shape irrespective of their colors, sizes or
Categories
Published
of 105
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Connectionist Account of Object-Substance distinction 1Running head: Connectionist Account of Object-Substance distinctionA Connectionist Account of the Object-Substance distinctionin Early Noun LearningEliana ColungaDepartment of Psychology1101 East Tenth Street, Bloomington, Indiana 47405-7007Linda B. SmithDepartment of Psychology1101 East Tenth Street, Bloomington, Indiana 47405-7007  Connectionist Account of Object-Substance distinction 2 Abstract Young children learning vastly different languages generalize nouns in the same systematic way.Young children extend the name of a solid object to other objects that match the exemplar in shapeirrespective of their colors, sizes or materials. However, when generalizing the name of a non-solidsubstance, children consider material to be more important. Thus, young children seem to knowthat objects and substances are fundamentally different and classified in different ways. Wheredoes this knowledgecome from? This paper presents evidence that this distinctionis a product of learned correlations among perceptual properties, syntactic cues, and the lexical category structureof early learned nouns. In nine experiments we show that (1) a connectionist network trained onnoun vocabularies that present the same statistical regularities characteristic of early learned nounsform a generalized distinctionbetween solids and non-solids;(2) the networks simulate bothglobal and finer-grained aspects of previous findings on children’s novel noun genralizations; and(3) the simulations generate new predictions that are supported by new experiments investigatingchildren’s novel noun generalizations.  Connectionist Account of Object-Substance distinction 3 A Connectionist Account of the Object-Substance distinction in EarlyNoun Learning An object-substancedistinction has played a role in discussions by linguists,philosophers,and psychologists(Prasada, 1993; Pelletier, 1979; Hall, 1996; Quine, 1960; Lucy, 1992; Piaget,1952; Xu, 1997). At its core, the distinctionis about discrete versus continuousquantities.Conceptually, objects are discrete and bounded unitary wholes, whereas substances are continuousunbounded masses. Whether an entity is considered an object or a substance determines itsgrammatical class in some languages, the categories one forms, the inferences one makes, and howone reasons about amount. This paper concerns a distinctionthat young children make that is notidentical to the object–substancedistinction,but that is arguably the developmental foundation forthose more mature and more abstract concepts.The distinctionchildren make is between solid and nonsolid things,between things that holdtheir shape when pushed and prodded and things that do not. These two distinctions oneconceptual and the other perhaps more perceptual do not map directly onto each other: there aresolid objects (table) and solid substances (wood), and also non-solid objects (bubble) and non-solidsubstances (water). However, in some ways, solidity seems an ideal expression of objectnessbecause solid things have stable bounded shapes, and similarly, non-solidityseems an idealexpression of substances since for non-solids material is stable and form is transient.The evidence certainly sugests that a solid-nonsoliddistinction is made early. In habituationtasks, young infants expect solids and non-solids to behave differently when transformed (Spelke,Breinlinger, Macomber, & Jacobson, 1992; Gibson, Owsley, Walker, & Megaw-Nyce, 1979; J,1979; Mash, Quinn, Dobson, & Narter, 1998). By the time children are 2 1/2-years-old, they alsoappear to know that solid and non-solid entities are fundamentally different kinds that arecategorized by different properties. Specifically, children expect solid things to be classified byshape and non-solid things to be classified by material (Soja, Carey, & Spelke, 1991). The  Connectionist Account of Object-Substance distinction 4literature indicates that childrens expectations about solid and non-solid categories are more robustin lexical categorization tasks, more robust for solids than for non-solids,emerge during the sameperiod that children learn names for common objects and substances and are different for childrenlearning different languages. The evidence also suggests that this distinctionmay be closely linkedto task and stimulus properties beyond solidity. The goal of this paper is to explain these data andthe developmental processes that give rise to childrens ideas about how categories of solids andnon-solids are organized, ideas that may be the starting point for a more conceptual understandingof objects and substances (Xu, 1997; Bloom, 1996, 2000). The phenomenon One task used to study children’s expectations about category organization is the NovelNoun Generalization task (Landau, Smith, & Jones, 1988; Soja et al., 1991). This task consists of showing the child an exemplar, labeling that exemplar, for example (e.g. “This is a dax.”), and thenasking the child to indicate what other things can be called by the same name. All the objects andnames used are novel, thus this task measures children’s expectations about how categories are tobe formed, rather than children’s knowledge about specific categories. The results indicate thatchildren expect categories of solid and non-solid things to be organized differently (Soja et al.,1991).In general, children generalize names for solids by shape and names for non-solids bymaterial. That is, there appears to be a shape bias for solid things and a material bias for non-solidthings. The shape bias for solids is a highly robust effect that has been replicated in many studiesof children’s novel noun generalizations (Landau et al., 1988; Soja et al., 1991; Samuelson &Smith, 1999; Imai & Gentner, 1997; Kobayashi, 1997). The material bias for non-solids is aweaker effect. Some experiments have found that children consistentlygeneralize names for anon-solid substances by material, but others have found that children are equally likely togeneralize names for non-solids by shape as by material (Soja et al., 1991; Samuelson & Smith,  Connectionist Account of Object-Substance distinction 51999; Imai & Gentner, 1997). However, all studies have found that children generalize names forsolids and non-solids differently; there is greater attention to the shapes of solids than non-solids,and greater attention to the materials of non-solids than solids. These results have been interpretedas indicating that children conceptualize solid things as discrete objects with fixed shapes andnon-solid things as materials with variable shapes (Soja et al., 1991, and others).Although this distinctionhas been widely replicated, the full literature contains a number of complexities that are potentiallyhighly informative of both the nature and srcin of this distinction.These facts comprise the specific data we seek to explain and they suggest that both languagelearning and the perceptual properties of the named entity matter —and that they do so in ways thatappear deeply related to the structure of the lexicon. There are two classes of effects that need to beexplained: task effects and history effects.Task effectsChildren’s attention to shape and material depends on a number of task factors including thelinguisticinput (the wording of the instructionsgiven to the children) and the perceptualcharacteristics of the stimuli involved. First, the task itself – whether the task is to name, to judgesimilarity or to group – matters. Children categorize solids according to shape and non-solidsaccording to material only in naming tasks. That is, after children are shown a solid object and toldits name, they systematically extend that name to other items of the same shape; however, if children are shown a solid object and asked to get another that is like it, they do not systematicallyselect shape matching objects (Landau et al., 1988; Soja et al., 1991; Samuelson & Smith, 1999).The fact that the shape bias for solids and the material bias for non-solids are specific to namingtasks suggests that the phenomenon has something to do with how lexical categories are organized.Second, within naming tasks, children’s generalizations can be shifted by syntactic cues(Soja, 1992). In English, count nouns are preceded by “a” or “an” and refer to discrete entities thatcan be counted (such as a cup or an airplane) and mass nouns are preceded by “some” or “much”
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks