High Frequency Words can Assist Language Acquisition
Frost, R. L. A, Monaghan, P., & Christiansen, M. H. (2016, June). Talk to be presented at the Fifth Implicit Learning Seminar, Lancaster, UK.
Learners can extract transitional information from speech and use it to infer word boundaries and linguistic regularities. Critically, studies suggest that statistical language learning may benefit from the presence of high-frequency marker words (Bortfeld et al., 26 2005) that may act as anchors around which speech segmentation can occur, while also assisting with grammatical categorisation (Monaghan & Christiansen, 2010).
To address these claims, we familiarised adults with a continuous stream of artificial speech comprising repetitions of 8 bisyllabic target words, and compared learning to the same language but with high-frequency monosyllabic marker words preceding target words. Critically, marker words distinguished target words into two distributionally-defined categories, which were otherwise unidentifiable. Participants completed a 2AFC test of segmentation, and a similarity judgement categorisation test containing word-pairs from the same versus different grammatical categories. We then tested transfer to a crosssituational word-action/object learning task, where target word categories were either consistent or inconsistent with the action/object distinction. This was then followed by a vocabulary test, which assessed whether participants had learnt the names of the actions and objects.
Participants in both training conditions segmented the speech stream better than chance, but only the marker word condition demonstrated evidence of categorisation on the transfer task. Data from the vocabulary task provided further evidence that categorisation was influenced by high frequency words: vocabulary learning for the marker word condition was significantly better for participants receiving consistent, rather than inconsistent, wordaction/object pairings. Findings indicate that high-frequency marker words may assist grammatical categorisation at the point at which speech segmentation is just being learnt.