The Role of Conceptual and Perceptual Information in Early Childhood Word Learning - Essay Example

The Role of Conceptual and Perceptual Information in Early Childhood Word Learning and the Shape Bias Controversy Introduction There's an ongoing debate on the role of conceptual information in early childhood word learning, on one hand the Attentional Learning Account holds that conceptual information begin to exert their influence only relatively late in the word learning. The essay will begin by focusing on the data that supports the second view and then that which supports the first view (The Attentional Learning Account view), then some criticism of the controversy itself and different views other than the two mentioned earlier will be presented, and finally a conclusion about the debate. The first part will briefly state the claims of the opponents of the Attentional Learning Account; the data from experiments they have conducted and their interpretation of experiments that are used as counter-arguments. The second part will discuss the claims of the proponents of the Attentional Learning Account view, and experiments conducted by them in support of their view. The third part will summarize some of the commentaries on the controversy and finally the conclusion will discuss which if any view is correct, and the importance of the controversy at hand. Part 1, In support of the view that conceptual information aids word learning childhood: In an experiment conducted by Booth and Waxman (2002), a group of three years old children were subject to the same objects which were separated into two groups, either as animate objects or artifacts. When some of the objects with eyes on them, which are strong perceptual cues to animacy, and the objects were presented as artifacts, the children characterized them as artifacts. This shows that even in the face of strong perceptual gestures, conceptual information still aids the process of word learning in childhood, a view that contradicts the Attentional Learning Account view. Something which sparked a number of responses and different interpretations to this experiment but which face certain difficulties that further support the view that conceptual information does have a role in early word learning. First it was noted that both conceptual and perceptual information affect the word learning process, in what sense then is perceptual information immune to conceptual information This simply means that if two factors affect the same variable, the only way one of the factors can be immune to the other is if one of them does not have an effect at all on the variable, something that the experiment mentioned above excludes. Second it was claimed by the Attentional Learning Account that perceptual information contributes directly to word learning unaided by any conceptual information; the experiment shows however that perceptual learning can be aided by conceptual information in a way that can change the outcome of the experiment. A special case of these two different views will be taken into account; namely the disagreement about the shape bias. The controversy over the role of conceptual learning in early word learning for children has initiated the shape bias controversy; briefly stated it is an implication of the two views outlined above. On the one hand the proponents of the Attentional Learning Account claim that shape bias does not emerge until infants are able to distinguish at least 50 nouns which is in line with the claim they support that conceptual knowledge is only available much later. On the other hand Booth and Waxman have shown with an experiment they conducted with 18 to 22 months old children with vocabulary that does not exceed 18 nouns that these children have extended their uses of words on the basis of shape. This experiment thus interpreted challenges the whole of the Attentional Learning Account proponents. In a paper in the Developmental Science journal, Booth and Waxman have shown experiments in support of the Attentional Learning Account view and have tried to show that they can be interpreted in a way that does not do any harm to their view on conceptual knowledge. In experiments conducted by Smith (2002) and Samuelson (2002) toddlers who were too young to have access to shape bias for standard in extending their usage of words, they were then trained to make use of shape for extending their nouns. At the end of the training however, these toddlers were able to extend the usage of nouns on the basis of shape. This is interpreted by the proponents of the Attentional Learning Account as evidence that with training, toddlers were able to detect correlations between novel nouns and shape based commonalities. Another interpretation is given, however, by the opponent view; as Booth and Waxman among others have interpreted the experiment differently, it does not necessarily show the Attentional Learning Account to be true. Even though toddlers could detect the correlations between nouns and shape after their training, Booth and Waxman gave two reasons to believe that this association was not the only mechanism underlying their performance. They first suggested that by the time toddlers entered the training they had already established a link between count nouns and object categories (Booth & Waxman, 2002a, 2003b; Waxman, 1999; Waxman & Booth, 2001, 2003; Waxman & Hall, 1993; Waxman & Markow, 1995, 1998). They also suggested that during the same time toddlers had their training, count nouns develop inductive forces in infants, which makes them look beyond perceptible properties of objects such as hidden even causal commonalities among them (Booth & Waxman, 2002a; Diesendruck & Bloom, 2003; S.A. Gelman & Kalish, 2006; Graham, Kilbreath & Welder, 2004; Welder & Graham,2001). Part 2, In support of the Attentional Learning Account: To paraphrase Smith and Johns in a paper titled "Naming in Young Children: a dumb Attentional Mechanism" previous research suggests that young children's naming is mainly a matter of mapping words to selected perceptual properties. They add that very young children can hear an object named once and generalize to more instances, a process in which they succeed more often than fail. In another experiment 2 to three years old children were given the names of certain objects and then asked which other objects had the same name; children based their generalizations on certain properties of objects. They see important features of the children's selections of properties, on which they based their generalizations; when children were subject to solid rigid bodies they used shape as a main property, however other properties were used when objects had eyes on them (for example texture or color). Another result that Smith and Johns see as important in this experiment is that children's chosen properties can be observed in named objects only; these properties were not used in non-named objects. This is important for the proponents of Attentional Learning Account because it shows that children's word interpretation is characterized by context specific selective attention. That means that the properties they choose to give attention to depend on the context in which the objects are given. This is important to show that shape is only one of the properties that attention might be drawn to depending on the context; one of the basic beliefs proponents of Attentional Learning Account hold. As Colunga and Smith point out; according to Attentional Learning Account opponents, knowledge is not a separated step of the mental life but it is allotted across the whole function that is consisted of perceiving, remembering, attending and acting (see Samuelson &Smith, 200.) This subtraction, however, was not well defined because it came from the opponent view form of (sense-think-act) idea. They also pointed out that On the other hand the proponents of Attentional Learning Account do not oppose conception itself but they oppose the sense-think-act definition of conception, for the proponents conception is a continuous function through all the steps of perceiving, attending, remembering and acting, not just a different step as sense-think-act which their opponents claim. In brief the partisans of sense-act approach think that the apprehension of children for concealed properties, their use of gradual event of decision making and .their long-term knowledge about regularities are all included in the processes of perceiving, remembering, attending and acting. Within this view there is some divergence pointed out by Soja, Carey and Spelke, between them and Landau Jones and Smith (hereafter LJS.) The disagreements they have are concerned with three main points, first the role of shape in children's projections of words referring to objects, second, the relation between word projection and word meaning and the third concern is; the proper interpretation of Quine. For example the first area of disagreement which is concerned with the role of shape in children's projections of words referring to objects is mainly about the perceptual basis of naming, in spite of the fact that the LJS account, which sees that count nouns referring to objects are projected on the basis of shape, is consistent with the LJS data, Soja, Carey and Spelke express doubts about whether the data captures the actual basis of the child's inductions concerning what or how words are used to name. The alternative view that they hold as opposed to LJS's is that count nouns referring to objects pick out of objects of the same kind. That means that children and adults show shape bias in the LJS studies not for the reason LJS suspect but rather because shape is often a good source of information about the object's kind. That is their new interpretation of the data available and obtained by LJS. Samuelson and Horst have taken a dynamic systems approach to the origin of shape bias. They have briefly stated the Attentional Learning Account, which they intend to extend, as a view that proposes that as children learn their early vocabulary, they learn a system of statistical regularities among linguistic devices, the properties of objects, and perceptual category organization. These learned associations then mechanistically to relevant properties of objects. That explains why children can learn new words so rapidly. According to the Attentional Learning Account as stated by Samuelson and Horst, then, children's attention to shape results from particular statistical structure of the linguistic context they have previously been exposed to. They maintain that "ALA (Attentional Learning Account) predicts individual cross cultural and task based differences in performance that are closely tied to the specifics of an individual history, the particular language being learned, and the specifics of the predictions supported in a growing body of work." Their work with the mechanisms of the decision children take when asked about a certain object's name or the category in which it falls; their experiment has shown, according to them, the importance of the timescale for our understanding of children's novel noun generalization behaviors. Part three, Commentaries on the Controversy: Writing about the future of the controversy, Keil have identified ten points of common ground, these ten claims are approved by both sides of the controversy; The first point of agreement was that young children will often extend word labels in ways that reflect the shape of the original taught on-object, which means that they can give different objects of the same kind the same name based on their shape. A second point of agreement is that young children do not extend their taught words on the basis of shape steadily; rather they sometimes use other attributes of objects. Both the opponents of the Attentional Learning Account and its proponents also seem to agree on a third point, which is that the naming patterns seem to change and develop in the way they are related to shape bias. They also agree that context in which an object is given can influence the behavior towards it, specifically it can influence labeling and to what degree the shape bias is manifested. The fifth point of agreement is that associative learning is real but also that there are non-associative components of human thought. Seventh; the shape bias is most obvious with bounded objects. An eighth point of agreement is that micro-genetic longitudinal studies including those that involve t raining can yield new insights. They are specifically useful to see how exposure to words and shape influences labeling and future word learning. The ninth point of agreement that Keil has pinned down was that both camps agree that it is reasonable to talk about perceptual and cognitive factors as influencing behavior. And the final point of agreement which Keil has noted was that they agree that experience prior to entering a study can make a difference to patterns of word learning. Keil notes that even with a common ground the disagreement have taken immunity to being solved, since all the date available from experiments is interpreted in a way that does not harm the theory by its proponents and in a way that contradicts it by its opponents. This, argues Keil, is the basic problem of both views, especially that of Attentional Learning Account that cannot be falsified. Conclusion: It is indeed important which view is correct; the disagreement cannot be dismissed as irrelevant, it can be however said that the debate is going in the wrong direction, since different interpretation of data, as we have seen, can amount to immunity of refutation for each view, and it is well known that immunity to refutation endangers the scientific status of any belief. A good step on the way to solving the controversy would be to define the data that can refute each view, and with the required experiments it can be shown which if any is the correct view. References: Booth, A. E. & Waxman, S. R. (2003) Bringing theories of word learning in line with the evidence. Cognition, 87, 215-218. Booth, A. E. & Waxman, S. R. (2008) Taking stock as theories of word learning take shape. Developmental Science, 11, 2, 185-19 Keil, F. C. (2008) The shape of things to come: The future of the shape bias controversy. Developmental Science, 11, 2, 216-218 Smith, L. B., Jones, S. S. & Landau, B. (1996) Naming in young children: A dumb Attentional mechanism Cognition, 60, 143-17 Colunga, E. & Smith, L. B. (2008) Knowledge embedded in process: the self-organization of skilled noun learning. Developmental Science, 11, 2, 195-203 Soja, N. N., Carey, S., Spelke, E.S. (1992) Perception, ontology and word meaning. Cognition, 45, 101-107 Samuelson, L. K. & Horst, J. S. (2008) Confronting complexity: insights from the details of behaviour over multiple timescales. Developmental Science, 11, 2, 209-215 Read More