Purchase this article with an account.
Katrina Ferrara, Sarah Furlong, Barbara Landau, Soojin Park; Detailed visual memory capacity is present early in life. Journal of Vision 2015;15(12):665. doi: 10.1167/15.12.665.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Previous studies have shown that adults can remember 2,500 images with great accuracy and fidelity (Brady et al., 2008). However, little is known about the extent of visual memory capacity early in life. Such detailed visual memory may be surprising, given the importance of generalization in domains such as word and concept learning (Jenkins et al., 2014). We adapted the method of Brady et al. to test the fidelity of children’s visual memory. Twenty-four 4-5 year-olds and twenty-four 6-7 year-olds were first shown 116 images of everyday objects (e.g., apple, ball), each presented for 3s. Children were then tested for recognition of familiar vs. new items. Three conditions varied the test pairings. In the Novel Category condition, familiar items were paired with categorically distinct items (e.g., a doll and a telephone). In Different Exemplar, familiar items were paired with items from the same category (e.g., two backpacks). In Different State, familiar items were paired with the same item in a different state (e.g., a closed container and the same container with an open lid). Children performed far above chance (50%) in all conditions: (4’s: 95% Novel Category, 94% Different Exemplar, 86% Different State; 6’s: 98%, 91%, 88%, respectively). For both ages, accuracy was lower for the Different State than the Novel Category condition. A subset of children (n = 10) were tested 18-21 days later and still performed above chance on the recognition test. The data show that children retain a great degree of visual detail from a large set of images; they can recall a specific category member and its particular state. This occurs spontaneously, as children were never instructed to remember the pictures. These results suggest that the ability to encode a high level of detail across a large number of images is well developed early in life.
Meeting abstract presented at VSS 2015
This PDF is available to Subscribers Only