Bejjanki, V. 1 , Randrup, E. 1 & Aslin, R. 2
1 Department of Psychology, Hamilton College, Clinton, NY, USA
2 Haskins Laboratories, New Haven, CT, USA
Efficient statistical learning can be viewed as an inference task: optimally integrating available sensory information with previously learned task-relevant regularities. Extensive research has shown that young children and infants can learn and utilize environmental regularities present in sensory information. However, it has also been reported that children younger than 8 years of age do not combine multiple sources of simultaneously available information in an optimal fashion. Thus, it remains unclear whether, and by what age, children combine sensory information with previously learned regularities in an adult manner. Here, we examine the performance of 6-7-year-old children when tasked with localizing a 'hidden' target by combining uncertain sensory information with regularities learned over repeated exposure to the task. We demonstrate that 6-7-year-olds learn and utilize task-relevant statistics in a statistically efficient fashion, and in a manner indistinguishable from adult behavior. We also show that variables such as task complexity can influence young children's behavior to a greater extent than that of adults, leading their behavior to look sub-optimal. These findings have important implications for how we should interpret failures in young children's ability to carry out sophisticated computations - these 'failures' need not be due to deficits in computational capacity.