Proceedings of the Annual Meeting of the Cognitive Science Society

As the use of naturalistic speech data of children's language experiences increases, the temporal dynamics of the speech environment becomes a more obvious aspect of speech to investigate. To connect the temporal dynamics in child- centered speech to existing experimental work showing that the temporal presentation of items has measurable effects on learning, it is important to develop measures that quantify temporal patterns in speech. The present work explores one such measure, the burstiness metric, and investigates word burstiness and its relationship with frequency, its behavior across different timescales, and whether it can be used to quantify constructs of massed and spaced orders. Our findings suggest that while related, word burstiness is not an index of frequency, that burstiness varies across both words of different lexical classes and timescales, and that it does not appropriately capture massed and spaced temporal patterns. We discuss implications for how this measure may be used for child- centered audio.