Comparing two new single-process accounts of the restriction of argument-structure generalizations
Lead Research Organisation:
University of Liverpool
Department Name: Psychology
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Organisations
Publications
Ambridge
(2011)
Child Language Acquisition: Contrasting Theoretical Approaches
Ambridge B
(2009)
A semantics-based approach to the "no negative evidence" problem.
in Cognitive science
Ambridge B
(2013)
The retreat from overgeneralization in child language acquisition: word learning, morphology, and verb argument structure.
in Wiley interdisciplinary reviews. Cognitive science
Ambridge B
(2010)
Children's judgments of regular and irregular novel past-tense forms: new data on the English past-tense debate.
in Developmental psychology
Ambridge B
(2012)
Semantics versus statistics in the retreat from locative overgeneralization errors.
in Cognition
Ambridge B
(2013)
How do children restrict their linguistic generalizations? An (un-)grammaticality judgment study.
in Cognitive science
Ambridge B
(2011)
Research Methods in Child Language - A Practical Guide
Ambridge B
(2013)
Experimental methods in studying child language acquisition.
in Wiley interdisciplinary reviews. Cognitive science
Ambridge B
(2012)
Avoiding dative overgeneralisation errors: semantics, statistics or both?
in Language, Cognition and Neuroscience
| Description | The aim of the project was to investigate competing explanations for how children come to avoid the production of ungrammatical utterances (e.g., *The funny joke giggled the man), whilst retaining the capacity to produce novel grammatical utterances that they have not previously encountered in exactly that form (a hallmark of human language). The project has made a significant contribution in this area, and has so far yielded five journal articles (three of which are already published or in press), with more to follow (exceeding our proposed output of three). The first study investigated whether two previously-proposed accounts were supported for overgeneralization errors involving the double-object dative construction (e.g., *John said Sue something). Under the entrenchment hypothesis, repeated presentation of a verb in permitted constructions (e.g., John said something to Sue) leads to the ever-strengthening inference that its use in non-attested constructions is not permitted. Under the semantic verb class hypothesis, speakers form classes of verbs with similar meanings that are restricted to particular constructions only. As predicted by the entrenchment hypothesis, the unacceptability of overgeneralization errors was significantly correlated with verb frequency. As predicted by the semantic verb class hypothesis, children used the meanings denoted by novel, experimentally-taught verbs to determine the constructions in which they can and cannot appear. Following positive reviews, a revised journal article has recently been resubmitted to the leading international linguistics journal (Language). The remaining studies compared two new single-process accounts designed to incorporate both entrenchment and semantic verb class effects. Under the two-level-entrenchment account, statistical entrenchment effects are extended to verbs with similar meanings. Under the incremental learning of verb and construction semantics account, speakers avoid extending verbs into constructions with which they are semantically incompatible. The findings of these studies supported the incremental learning of verb and construction semantics account. In particular, we found that independently manipulating the semantics of either the construction (Ambridge et al, 2009; Cognitive Science) or the verb (Ambridge et al, in press; Cognitive Linguistics) influences the perceived grammaticality of transitive causative uses of novel verbs (e.g., *The man meeked the ball). We have also completed an analagous study on the double-object dative construction, which is currently in the process of being written up. We have also conducted three additional studies that go beyond those outlined in the initial proposal: 1) An extension of the Language study discussed above that investigates overgeneralizations involving the locative constructions (e.g., *I spilled the floor with juice; *I filled some juice into the cup). This study is currently in the process of being written up. 2) An extension of the semantics studies discussed above to overgeneralizations involving un-prefixation (e.g., *I unclosed the door). This study is particularly important as it extends both the theoretical account and the grammaticality judgment methodology from the domain of syntax into the domain of morphology, and has been submitted to Cognition. 3) An extension of the incremental learning of verb and construction properties account to the domain of the English past-tense, where the relevant properties are phonological, as opposed to semantic. This study is currently in press at Developmental Psychology. |
| Exploitation Route | Extend to other languages (I plan to apply for a grant to do just this). |
| Sectors | Education |
| URL | http://www.benambridge.com |
| Description | As with most research projects that seek to understand the processes behind typical language development, the short term impacts are mostly scientific. However, in the longer term, research on the formation and restriction of linguistic generalizations will inform interventions for children with language impairment, and educational practice more generally. This research also suggests useful strategies for adult learners of a second language (whether this is English or another language). In the longer term, our findings regarding the acquisition of the ability to form and restrict linguistic generalizations will benefit clinicians working with children with language impairment; we have ensured that these findings will be communicated to this group by presenting our findings in journals/conferences that are widely read/attended by language practitioners. |
| First Year Of Impact | 2010 |
| Description | "Retreat" from argument-structure overgeneralization errors as a consequence of the acquisition of verb and construction semantics. |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Other academic audiences (collaborators, peers etc.) |
| Results and Impact | Ambridge, B. (2009) "Retreat" from argument-structure overgeneralization errors as a consequence of the acquisition of verb and construction semantics. Paper presented at the biennial meeting of the Society for Research in Child Development, April 2009, Denver: CO, USA. NA |
| Year(s) Of Engagement Activity | 2009 |
| Description | Semantic verb classes and the retreat from overgeneralization error |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Other academic audiences (collaborators, peers etc.) |
| Results and Impact | Ambridge, B., Pine, J.M. & Rowland, C.F. (2010). Semantic verb classes and the retreat from overgeneralization error. Paper presented at the Boston University Conference on Language Development, November 2010. Boston, MA, USA Increase in awareness of my research, and of citations |
| Year(s) Of Engagement Activity | 2010 |
| Description | Testing a new account of the acquisition of linguistic constructions |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Other academic audiences (collaborators, peers etc.) |
| Results and Impact | Ambridge, B. (2010). Testing a new account of the acquisition of linguistic constructions. Conference of the Linguistics Association of Great Britain, September, 2010, Leeds, UK. NA |
| Year(s) Of Engagement Activity | 2010 |
| Description | Testing a new account of the acquisition of linguistic constructions - Competing Motivations |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Other academic audiences (collaborators, peers etc.) |
| Results and Impact | Ambridge, B. (2010). Testing a new account of the acquisition of linguistic constructions. Paper presented at the conference on Competing motivations in language acquisition, November, 2010, Leipzig, Germany. NA |
| Year(s) Of Engagement Activity | 2010 |
| Description | The no-negative-evidence problem for children's argument-structure overgeneralization errors. Why statistical entrenchment is not sufficient (and pre-emption isn't necessary). |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Other academic audiences (collaborators, peers etc.) |
| Results and Impact | Ambridge (2008). The no-negative-evidence problem for children's argument-structure overgeneralization errors. Why statistical entrenchment is not sufficient (and pre-emption isn't necessary). Paper presented in the symposium Solving the no-negative evidence problem using positive evidence: Data from mathematical, computational, elicited-production and grammaticality-judgement studies at the 11th meeting of the International Association for the Study of Child Language, July 2008, Edinburgh. NA |
| Year(s) Of Engagement Activity | 2008 |