Comparing generalisation in children and adults learning an artificial language
Elizabeth Wonnacott, Helen Brown, Kate Nation · 2013
Successful language acquisition involves generalization, but learners must balance this against the acquisition of lexical constraints. Examples occur throughout language. For example, English native speakers know that certain noun-adjective combinations are impermissible (e.g., strong winds, high winds, strong breezes, *high breezes). Another example is the restrictions imposed by verb sub-categorization (e.g., I gave/sent/threw the ball to him; I gave/sent/threw him the ball; I donated/carried/pushed the ball to him; * I donated/carried/pushed him the ball; Baker, 1979). A central debate has been the extent to which learning such patterns depends on semantic cues (Pinker, 1989) and/or distributional statistics (Braine et al., 1990). The current experiments extend previous work which used Artificial Language learning to demonstrate that adults (Wonnacott et al., 2008) and 6 year olds (Wonnacott, 2011) are able to learn lexically based restrictions on generalization using distributional statistics. Here we directly compare the two age groups learning the same artificial language, with a view to exploring maturational differences in language learning. In addition to manipulating frequency (across high and low frequency items) and quantity of exposure (across days), languages were constructed such that a word’s semantic class was helpful for learning the restrictions for some types of lexical items, but potentially misleading for others.