RT Journal Article
SR Electronic
T1 The development of Bayesian integration in sensorimotor estimation
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 136267
DO 10.1101/136267
A1 Claire Chambers
A1 Taegh Sokhey
A1 Deborah Gaebler-Spira
A1 Konrad Kording
YR 2017
UL http://biorxiv.org/content/early/2017/05/12/136267.abstract
AB If the brain is inherently Bayesian, then behavior should show the signatures of Bayesian computation from an early stage in life without the need for learning. Children should integrate probabilistic information from prior and likelihood distributions to reach decisions and should be as statistically efficient as adults. To test this idea, we examined the integration of prior and likelihood information in a simple position estimation task comparing children aged 6-11 years and adults. During development, estimation performance became closer to the statistical optimum. Children use likelihood information as well as adults but are limited in their use of priors. This finding suggests that Bayesian behavior is not inherent but learnt over the course of development.