Children learn words through an accumulation of interactions grounded in context. Although many factors in the learning environment have been shown to contribute to word learning in individual studies, no empirical synthesis connects across factors. We introduce a new ultradense corpus of audio and video recordings of a single child's life that allows us to measure the child's experience of each word in his vocabulary. This corpus provides the first direct comparison, to our knowledge, between different predictors of the child's production of individual words. We develop a series of new measures of the distinctiveness of the spatial, temporal, and linguistic contexts in which a word appears, and show that these measures are stronger predictors of learning than frequency of use and that, unlike frequency, they play a consistent role across different syntactic categories. Our findings provide a concrete instantiation of classic ideas about the role of coherent activities in word learning and demonstrate the value of multimodal data in understanding children's language acquisition.word learning | language acquisition | multimodal corpus analysis | diary study A dults swim effortlessly through a sea of words, recognizing and producing tens of thousands every day. Children are immersed in these waters from birth, gaining expertise in navigating with language over their first years. Their skills grow gradually over millions of small interactions within the context of their daily lives. How do these experiences combine to support the emergence of new knowledge? In our current study, we describe an analysis of how individual interactions enable the child to learn and use words, using a high-density corpus of a single child's experiences and novel analysis methods for characterizing the child's exposure to each word.Learning words requires children to reason synthetically, putting together their emerging language understanding with their knowledge about both the world and the people in it (1, 2). Many factors contribute to word learning, ranging from social information about speakers' intentions (3, 4) to biases that lead children to extend categories appropriately (5, 6). However, the contribution of individual factors is usually measured either for a single word in the laboratory or else at the level of a child's vocabulary size (4, 6, 7). Although a handful of studies have attempted to predict the acquisition of individual words outside the laboratory, they have typically been limited to analyses of only a single factor: frequency of use in the language the child hears (8, 9). Despite the importance of synthesis, both for theory and for applications like language intervention, virtually no research in this area connects across factors to ask which ones are most predictive of learning.Creating such a synthesis, our goal here, requires two ingredients: predictor variables measuring features of language input and outcome variables measuring learning. Both of these sets of measurements can be problematic.Examining predict...