Young children seem to overextend a 'seeing = knowing rule' so that they ne
glect to notice that people gain knowledge from inferring as well as from s
eeing. Yet that cannot be a sufficient explanation of children's problems w
ith conceptualizing knowledge, because rule underextension occurs in childr
en's claims to know something that they have not seen (nor inferred). The t
wo errors were studied together with pairs of children aged 4 and 5 years.
Each child had their own box, items on the table were shared out into the b
oxes, and either both children, or neither, or one of them, looked in their
own box. Children were asked if they and the other knew what was in each o
ther's box; and were asked for explanations and predictions. About a quarte
r of the children showed full competence. In others, overextension and unde
rextension occurred; yet almost all children explained that inference was i
nvolved in knowing, without bias towards giving such explanation more for s
elf than for other. Error-patterns were not predictable from a test for und
erstanding the term 'know'. It is suggested that children have a framework
conception of 'knowing' in which another's mind is treated as similar co ow
n mind, but problems arise in implementing that insight before children man
age to conceptualize constraints on knowledge.