Marcus

Rule Learning by Seven-Month-Old Infants

Introduction:

One learning mechanism that young infants can exploit is statistical in nature. […] For example, transitional probabilities - estimates of how likely one item it is to follow another.
It has been suggested that mechanisms that track statistical information […] may suffice for language learning. The alternative possibility considered here is that children might possess at least two learning mechanisms, one for learning statistical information and another one for learning “algebraic rules” - open-ended abstract relations for which we can substitute arbitrary items.
To date, there has been no direct empirical test for determining whether young infants can actually learn simplified versions of such algebraic rules.

Conclusions:

Our results do not call into question the existence of statistical learning mechanisms, but show that such mechanisms do not exhaust the child's repertoire of learning mechanisms.
A system that was only sensitive to transitional probabilities between words, could not account for any of these results, because all the words in the test are novel […]
Similarly, a system that noted discrepancies with stored sequences of words could not account for the results […] because both the consistent and inconsistent items differ from any stored sequences of words.

We propose that a system that could account for our results is one in which infants extract abstract algebra-like rules that represent relationships between placeholders (variables), such as “the first item X is the same as the third item Y”. If our position is correct, then infants possess at least two distinct tools for learning about the world and attacking the problem of language learning:
1. one device that tracks statistical relationships such as transitional probabilities;
2. and another one that manipulates variables, allowing children to learn rules.