Adults are experts at using visual input to recognize objects, like tools, animals, books, or other common items, and at recognizing materials, like cloth, wood, or plastic. While we know a great deal about how children learn to recognize objects, we know very little about how they learn to recognize materials. This project will help us understand how children's abilities to categorize real-world materials change during childhood, and will establish what information children use for recognition as they get older. Understanding material categorization may illuminate many similar tasks that depend on the same kinds of visual computations. Additional work in this domain will likely reveal important links between material categorization and a wide range of other processes including visual search and visual attention. The abilities that probably contribute to material recognition, such as texture perception, continue to develop until about 10 years of age, but at present, we do not know what is changing. What mechanisms allow children to recognize materials at different ages?

This project will focus on understanding how children at different ages use pure texture information and global image patterns to recognize materials. A sophisticated computer graphics model will make it possible to use images that represent real textures to create stimulus displays that represent artificial textures; the use of artificial textures that have some information missing and other information intact will reveal what features 5- to 10-year-old children rely upon to recognize materials. Children in these experiments will be asked to categorize presented images by assigning labels to them using a touchscreen interface (naming images as "Wood," "Stone," or "Plastic"), or will be asked to indicate which of several target images matches a sample image according to the material pictured in the images. Manipulating image appearance using a texture synthesis algorithm will permit evaluation of how children's categorization and matching of natural materials is affected by what information is available.

Agency
National Science Foundation (NSF)
Institute
Division of Behavioral and Cognitive Sciences (BCS)
Type
Standard Grant (Standard)
Application #
1727427
Program Officer
Peter Vishton
Project Start
Project End
Budget Start
2017-08-01
Budget End
2020-07-31
Support Year
Fiscal Year
2017
Total Cost
$149,744
Indirect Cost
Name
North Dakota State University Fargo
Department
Type
DUNS #
City
Fargo
State
ND
Country
United States
Zip Code
58108