Would LuLu call this a blicket? Categorization using a yes/no task with four year olds
¹Bentley Gibson, ² Dr. Laura Namy and Jessica Hammock
¹Spelman College, Atlanta, GA;
²Emory University, Atlanta GA

Abstract

Previous research suggests that comparison may play a role in children’s object categorization. When four-year-olds only receive one example of an object, they tend to use shallow, perceptual characteristics (e.g., shape, color) to form categories. However, after comparing multiple objects within a category, children are more likely to categorize according to conceptual commonalities (e.g., function). The purpose of this study is to clarify the role of comparison for children’s object categorization by using a “yes/no” task to determine whether children would accept both perceptual and conceptual matches as category members if given the opportunity to do so. Children were assigned to either a Compare or a No Compare condition. Those in the Compare condition learned a novel label (e.g., “blicket”) for two perceptually similar target objects from the same object category (e.g., an apple and a pear). Those in the No Compare condition learned a novel label for only one of the target objects. The experimenter then presented a series of four choice pictures including a perceptually distinct taxonomic match, a perceptually similar out of kind object, a perceptually similar category member, and an unrelated distracter. For each, the experimenter asked the child whether it belonged in the target category. Overall, children were generally conservative categorizers in this task, extending the category label consistently only to the “both” choice. However, those in the compare condition were significantly more likely to include the taxonomic matched object than those in the no compare condition. This study replicates the effect of comparison on categorization and confirms that children’s use of non-obvious conceptual properties as a basis for categorization increases following comparison of two or more category instances.

Introduction

• Young children often use perceptual properties like shape as a basis for categorizing objects.
• Previous studies show that comparison across items helps children look beyond the perceptual and arrive at the conceptual understanding of categories. (Gentner & Namy, 1999; Namy & Gentner,2002).
• The studies that examine this phenomena have used a forced choice task to compare perceptual to conceptual matching, but the two often work together. (Gentner & Namy, 2006).
• Using the yes/no task allows us to clarify how comparison influences children’s categorization strategies.

Methods and Materials

Questions of Interest

1. Do children show priority for perceptual versus conceptual (taxonomic) bases for classifying objects when given a “yes/no” task that allows them to employ both bases for categorizing objects?
2. Do these patterns vary when given to opportunity to compare across category instances?

Procedure

• The stimuli consists of laminated photographs of familiar objects to preschoolers
• 10 categories of objects were labeled using novel words (e.g., “blicket”)
• A stuffed ladybug named LuLu was used to teach the children the novel terms.

Results

• A mixed design ANOVA indicated a marginal interaction between choice type and condition, F(3,84)=2.173, p=.097.
• Children in the compare condition said yes to the taxonomic choice more than those in the no compare group.
• As expected, comparison to chance revealed that children overwhelmingly accepted the both choice and rejected the neither options
• Those in the no compare condition accepted both the taxonomic and perceptual choices reliably less often than expected by chance
• Those in the compare condition accepted the perceptual choice at below chance rates but were at chance in accepting the taxonomic choice.

Conclusions and Future Studies

Children are conservative categorizers, preferring to include only those objects that share both taxonomic and perceptual commonalities.

Comparison increases children’s willingness to include in their categories objects that are taxonomically but not perceptually related.

This study helps further explain the comparison effect observed using the forced choice task, confirming that comparison encourages children to move beyond perceptual similarity as a basis for categorization.

Given what we have learned about the utility of comparison and how it helps children categorize familiar objects, future studies will examine if also facilitates children’s ability to incorporate novel category instances (e.g., starfruit) into their existing categories (e.g., fruit)

Resources

This project was supported by the Howard Hughes Medical Institute. I also thank the SURE Program for their support, Dr. Laura Namy for her mentorship, and Jessica Hammock for her guidance.

References

1. Gentner, D., & Namy, L. (2006) Analogical processes in language learning. In Current Directions in Psychological Science, 15(6), 297-302.
2. Namy, L. & Gentner, D. (2002). Making a silk purse out of two sow’s ears: Young children’s use of comparison in category learning. Journal of Experimental Psychology, 131(1), 5-15.
3. Namy, L. & Gentner, D. (1999). Comparison in the development of categories. Cognitive Development, 14, 487 – 501.
4. Smith, L. (2005) Action alters shape categories. Cognitive Science, 29(4), 665-679.