Distributional category learning by 12-month-old infants: an investigation into the role of prosody and distributional frames

Distributional information is a potential cue for learning syntactic categories. Recent artificial grammar studies demonstrate sophisticated distributional learning by young infants. Here we investigate the possible mechanisms and representations underlying this ability. Does prosody constrain distributional analysis? What specific distributional relations do learners track? Twelve-month-old infants were exposed to an artificial language comprised of 3-word-sentences of the form aXb and cYd, where X and Y words differed in the number of syllables. Subsequently they were tested on novel utterances that were consistent or inconsistent with the training sentences. In Experiment I, infants showed evidence for having learned the relevant relations by successfully discriminating between novel grammatical and ungrammatical sentences. In Experiment II, we asked whether prosody influences infants’ distributional analysis. Contrary to our expectations, infants did not show a preference for relations between words that fell within a prosodic unit over those that straddled a prosodic boundary. In Experiment III, we explored whether infants’ success in the first experiment arose from their representation of nonadjacent relations or distributional frames. Our results did not support a frames hypothesis. We discuss these results and offer hypotheses regarding the nature of infants’ distributional learning abilities.