Affiliation:
1. University of California Davis
2. Princeton University
Abstract
AbstractAdults learning a new language tend to judge unconventional utterances more leniently than fluent speakers do; ratings on acceptable utterances, however, tend to align more closely with fluent speakers. This asymmetry raises a question as to whether unconventional utterances can be statistically preempted by conventional utterances for adult learners. We report a preregistered study that provided undergraduates in Spanish classes with three days of exposure to conventional Spanish sentences without feedback. Judgment data reveal a significant effect of statistical preemption, particularly on intermediate learners, as predicted: Repeatedly witnessing conventional sentences led learners to subsequently judge as significantly lower the corresponding unconventional formulations in comparison to unrelated unconventional sentences. Current findings indicate that adult learners can take advantage of statistical preemption to learn the unacceptability of unconventional sentences from repeated exposure to acceptable alternatives, without explicit instruction or feedback.