ABSTRACT
According to frequency-based approaches to acquisition, novel constructions are initially acquired through exposure to low-variability input and the engagement of cognitive mechanisms, such as fast mapping and category formation. However, previous second language (L2) studies have reported considerable variation in learners' ability to extract novel patterns when listening to sentences created with low-variability input. For example, in a series of studies involving the Esperanto transitive construction, few L2 participants could extract its key morphological (accusative case marking) and structural (variable word order) features, despite a variety of experimental manipulations to enhance its salience. Synthesizing the findings from these prior Esperanto experiments, this chapter compares the effectiveness of experimental manipulations involving input distribution, learner characteristics, and several aspects of the learning context to determine which factors were associated with novel pattern learning. The findings are discussed in light of several salience variables which might facilitate L2 morphosyntactic learning.
