Statistical learning of new visual feature combinations by infants

J. Fiser, Richard N. Aslin

Research output: Contribution to journalArticle

302 Citations (Scopus)

Abstract

The ability of humans to recognize a nearly unlimited number of unique visual objects must be based on a robust and efficient learning mechanism that extracts complex visual features from the environment. To determine whether statistically optimal representations of scenes are formed during early development, we used a habituation paradigm with 9-month-old infants and found that, by mere observation of multielement scenes, they become sensitive to the underlying statistical structure of those scenes. After exposure to a large number of scenes, infants paid more attention not only to element pairs that cooccurred more often as embedded elements in the scenes than other pairs, but also to pairs that had higher predictability (conditional probability) between the elements of the pair. These findings suggest that, similar to lower-level visual representations, infants learn higher-order visual features based on the statistical coherence of elements within the scenes, thereby allowing them to develop an efficient representation for further associative learning.

Original languageEnglish
Pages (from-to)15822-15826
Number of pages5
JournalProceedings of the National Academy of Sciences of the United States of America
Volume99
Issue number24
DOIs
Publication statusPublished - Nov 26 2002

Fingerprint

Learning
Aptitude
Complex Mixtures
Observation

ASJC Scopus subject areas

  • Genetics
  • General

Cite this

Statistical learning of new visual feature combinations by infants. / Fiser, J.; Aslin, Richard N.

In: Proceedings of the National Academy of Sciences of the United States of America, Vol. 99, No. 24, 26.11.2002, p. 15822-15826.

Research output: Contribution to journalArticle

@article{ca4bd316c4a149da924547cf04bd043c,
title = "Statistical learning of new visual feature combinations by infants",
abstract = "The ability of humans to recognize a nearly unlimited number of unique visual objects must be based on a robust and efficient learning mechanism that extracts complex visual features from the environment. To determine whether statistically optimal representations of scenes are formed during early development, we used a habituation paradigm with 9-month-old infants and found that, by mere observation of multielement scenes, they become sensitive to the underlying statistical structure of those scenes. After exposure to a large number of scenes, infants paid more attention not only to element pairs that cooccurred more often as embedded elements in the scenes than other pairs, but also to pairs that had higher predictability (conditional probability) between the elements of the pair. These findings suggest that, similar to lower-level visual representations, infants learn higher-order visual features based on the statistical coherence of elements within the scenes, thereby allowing them to develop an efficient representation for further associative learning.",
author = "J. Fiser and Aslin, {Richard N.}",
year = "2002",
month = "11",
day = "26",
doi = "10.1073/pnas.232472899",
language = "English",
volume = "99",
pages = "15822--15826",
journal = "Proceedings of the National Academy of Sciences of the United States of America",
issn = "0027-8424",
number = "24",

}

TY - JOUR

T1 - Statistical learning of new visual feature combinations by infants

AU - Fiser, J.

AU - Aslin, Richard N.

PY - 2002/11/26

Y1 - 2002/11/26

N2 - The ability of humans to recognize a nearly unlimited number of unique visual objects must be based on a robust and efficient learning mechanism that extracts complex visual features from the environment. To determine whether statistically optimal representations of scenes are formed during early development, we used a habituation paradigm with 9-month-old infants and found that, by mere observation of multielement scenes, they become sensitive to the underlying statistical structure of those scenes. After exposure to a large number of scenes, infants paid more attention not only to element pairs that cooccurred more often as embedded elements in the scenes than other pairs, but also to pairs that had higher predictability (conditional probability) between the elements of the pair. These findings suggest that, similar to lower-level visual representations, infants learn higher-order visual features based on the statistical coherence of elements within the scenes, thereby allowing them to develop an efficient representation for further associative learning.

AB - The ability of humans to recognize a nearly unlimited number of unique visual objects must be based on a robust and efficient learning mechanism that extracts complex visual features from the environment. To determine whether statistically optimal representations of scenes are formed during early development, we used a habituation paradigm with 9-month-old infants and found that, by mere observation of multielement scenes, they become sensitive to the underlying statistical structure of those scenes. After exposure to a large number of scenes, infants paid more attention not only to element pairs that cooccurred more often as embedded elements in the scenes than other pairs, but also to pairs that had higher predictability (conditional probability) between the elements of the pair. These findings suggest that, similar to lower-level visual representations, infants learn higher-order visual features based on the statistical coherence of elements within the scenes, thereby allowing them to develop an efficient representation for further associative learning.

UR - http://www.scopus.com/inward/record.url?scp=0037180434&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037180434&partnerID=8YFLogxK

U2 - 10.1073/pnas.232472899

DO - 10.1073/pnas.232472899

M3 - Article

VL - 99

SP - 15822

EP - 15826

JO - Proceedings of the National Academy of Sciences of the United States of America

JF - Proceedings of the National Academy of Sciences of the United States of America

SN - 0027-8424

IS - 24

ER -