Abstract
Learning to produce sentences involves learning patterns that enable the generation of new utterances. Language contains both verb-specific and verb-general regularities that are relevant to this capacity. Previous research has focused on whether one source is more important than the other. We tested whether the production system can flexibly learn to use either source, depending on the predictive validity of different cues in the input. Participants learned new sentence structures in a miniature language paradigm. In three experiments, we manipulated whether individual verbs or verb-general mappings better predicted the structures heard during learning. Evaluation of participants’ subsequent production revealed that they could use either the structural preferences of individual verbs or abstract meaning-to-form mappings to construct new sentences. Further, this choice varied according to cue validity. These results demonstrate flexibility within the production architecture and the importance of considering how language was learned when discussing how language is used.
Highlights
As speakers, we do not repeat the same set of sentences every day but produce new sentences to describe events in the world as well as our internal thoughts
The proportion of PA order [=PA responses/(PA responses + AP responses)∗100] for each verb bias condition is shown in Figure 2 (PA order proportion for AP-only = 18.66, Syn-AP = 27.26, Alt = 32.89, PA-only = 42.22)
Participants’ descriptions of new events, which could not rely on memorization, showed that the likelihood of PA order varied according to verb bias
Summary
We do not repeat the same set of sentences every day but produce new sentences to describe events in the world as well as our internal thoughts. A persistent question for research is whether the sentence production system accomplishes this feat by preferentially encoding and using knowledge about how specific verbs are used in different types of sentences or knowledge about how broad meanings are mapped to different word orders independent of specific verbs. We use a cue validity framework to test the idea that the mental architecture for learning to produce sentences might flexibly adapt to use either type of information depending on statistical regularities in the input. The focus of this paper is grammatical encoding (Levelt, 1989). This is the stage at which different words are assigned to different grammatical functions (e.g., subject, direct object) and assembled into an ordered
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
