Observing and Modeling Developing Knowledge and Uncertainty During Cross-Situational Word Learning

Being able to learn word meanings across multiple scenes consisting of multiple words and referents (i.e., cross-situationally) is thought to be important for language acquisition. The ability has been studied in infants, children, and adults, and yet there is much debate about the basic storage and retrieval mechanisms that operate during cross-situational word learning. It has been difficult to uncover the learning mechanics in part because the standard experimental paradigm, which presents a few words and objects on each of a series of training trials, measures learning only at the end of training, after several occurrences of each word-object pair. Diverse models are able to match the final level of performance of the standard paradigm, while the rich history and context of the learning trajectories remain obscured. This paper examines accuracy and uncertainty over time in a version of the cross-situational learning task in which words are tested throughout training, as well as in a final test. With similar levels of performance to the standard task, we examine how well the online response trajectories match recent hypothesis- and association-based computational models of word learning.