Words can slow down category learning

Brojde, Chandra L.; Porter, Chelsea; Colunga, Eliana

doi:10.3758/s13423-011-0103-z

Cited by 13 publications

(36 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(Smith et al, 2002). Evidence from adults suggests that redundant labels facilitate category learning (Lupyan et al, 2007) possibly because they increase selective attention to category relevant dimensions (Brojde et al, 2011). Thus increased accuracy for between-category relative to within-category discriminations.…”

Section: Chapter Viii: General Discussionmentioning

confidence: 99%

“…These results could be taken as evidence for the idea that labels direct attention to specific dimensions, but that when attention is already biased to a given dimension, this increase in selective attention to that dimension makes it quite difficult to attend instead to another dimension (cf. Brojde et al, 2011). Additionally, the fact that size preferrers were equally good at learning categories organized by their preferred and nonpreferred dimension without a label, but were slightly slower to learn size categories with the label, could suggest that size is a harder dimension to attend to in category learning.…”

Section: Chapter Viii: General Discussionmentioning

confidence: 99%

“…On the one hand, that the brightness preferrers would have increased difficulty for learning size categories would seem to support the idea that labels can increase our existing attentional biases making it even harder to switch attention to another dimension (cf. Brojde et al, 2011). However, this would not explain why the size preferrers were also slower to learn size categories in the label experiment-unless both preferrer groups have some bias to attend to brightness that was exaggerated by the presence of a redundant label.…”

Section: Category Learningmentioning

confidence: 91%

“…If this is the case, it would suggest that labels enhance category representations allowing children to focus on category-relevant dimensions. However, recent work on adults' category learning suggests that when learners are biased to attend to a specific dimension, labels increase this bias making it more difficult to learn categories organized by another dimension (Brojde, Porter, & Colunga, 2011). Thus, it is possible that labels could make it even more difficult for brightness preferrers to selectively attend to size in the category learning task.…”

Section: Methodsmentioning

confidence: 99%

“…If the latter is true, then not all category learning should benefit from the presence of labels (cf. Brojde et al, 2011). In particular, learning holistic categories, which require representing overall similarity rather than selecting specific dimensions, requires less selective attention Ashby & Maddox, 2011;Sloutsky, 2010).…”

Section: What Do Labels Really Do?mentioning

confidence: 99%

See 4 more Smart Citations

The role of word learning in the development of dimensional attention

Perry¹

View full text Add to dashboard Cite

Previous work shows that young children focus on holistic (or overall) similarity and older children focus on dimensional similarity (selectively attending to one property to the exclusion of others). Research on early word learning, however, suggests that process of learning new words trains attention towards category-relevant dimensions via regularities in the linguistic and physical environment. Thus, over development, children learn to attend to specific dimensions when making nominal category judgments-they selectively attend to shape, for example, when learning names for solid objects. In four experiments, I asked a question fundamental to our understanding of dimensional attention: does word learning scaffold attention to dimensional similarity in more general contexts. The results of Experiment 1 showed that children who are holistic classifiers are slower than dimensional classifiers to learn categories of objects that vary along both a category-relevant dimension (e.g. size) and a category-irrelevant dimension (e.g. brightness). However, the results of Experiment 2 showed that when children were presented with incidental labels during category learning, holistic classifiers learn the categories as quickly as dimensional classifiers. In a follow-up similarity classification task, children who had been holistic classifiers showed an increase in dimensional attention only if they had been in the label experiment. In Experiments 3 and 4, I examined category learning with and without a label in children who preferred to selectively attend to one dimension of similarity (e.g. brightness) regardless of whether this means selecting dimensional or holistic matches in a classification task. The results of these experiments provide a more complete picture xi TABLE OF CONTENTS LIST OF TABLES .

show abstract

Section: Chapter Viii: General Discussionmentioning

confidence: 99%

Section: Chapter Viii: General Discussionmentioning

confidence: 99%

Section: Category Learningmentioning

confidence: 91%

Section: Methodsmentioning

confidence: 99%