Morphological analysis and learner interlanguage
Time: 01:00pm - 02:00pm
Place: Ballantine Hall 141
Scott LedbetterThis research describes a system designed to automatically analyze and interpret learner language. A longitudinal corpus of learner Hungarian was collected and annotated for errors, including features of linguistic importance. These features allow for an approximation of the interlanguage of a learner or a group of learners, which can possibly indicate things like the relative stage of acquisition for a given phenomenon and, perhaps, for the language as a whole. In this talk, we'll present our progress with the ongoing work in building the corpus and the analyzer, as well as possible applications and their evaluation.
In category: Computational linguistics
Why Accusative A? Topic Highlighting in Argentine Spanish
Time: 12:15pm - 01:15pm
Place: BH 006
Mark HoffDifferential Object Marking (DOM) is represented in Spanish by the a-marking of certain direct objects. Much attention has been paid to DOM of animate objects with a multitude of studies examining its variable use, but few authors address the a-marking of inanimates. Tippets (2011) makes brief mention of the phenomenon in his Buenos Aires data, adding that this dialect exhibits frequent and often unexpected a-marking. Weissenrieder (1985) and others discuss cases where inanimates are a-marked to disambiguate subject and direct object or where a-marking occurs with certain specific verbs. Here I examine in detail the "non-normative" DOM of inanimates in Argentine Spanish. I argue that this marking serves as a pragmatic tool to highlight a direct object’s topicality in discourse. Examples from the television series Farsantes and a variety of textual sources are presented to support this claim and show the frequency and domains of use of this marking.
In category: Morphosyntax and semantics
Semantic Models and their Applications to Vocabulary Development in Children with Cochlear Implants and Normal Hearing Children
Time: 01:30pm - 02:30pm
Place: Psychology 128
Jon WillitsI will give a talk that has two parts. In the first half I will discuss semantic memory models and some of their applications to vocabulary development in typically developing children. I will show that these models can help predict interesting facts about language development, such as which words and concepts are easy and difficult to learn. The models also make suggestions about some underlying facts about the language learning process, such as the importance of different kinds of information. In the second half I will discuss some preliminary work on applying these models to language development with children with cochlear implants, such as modeling behavior on verbal fluency tasks.
In category: Child language acquisition
The Parser as Language Acquisition Device (PLAD) in L2 acquisition
Time: 02:30pm - 04:00pm
Place: Ballantine Hall 215
Laurent Dekydtspotter & Claire RenaudThis talk will discuss a hypothesis about the nature of second language (L2) development introduced in forthcoming work on the role of the parser in L2 grammatical acquisition. It is well-known that parsing (the analysis of sentences in real time) could play a role in grammatical acquisition (Fodor, 1998). The parser could be doing the detection, providing the triggers for change. The hypothesis examined here claims that the parser functions as the Language Acquisition Device in L2 acquisition. This possible mechanism relies on parsing as combining the generation of Universal Grammar-constrained structures with grammatical licensing by a (functional) lexicon at each step of processing. The grammatical specifications for a new language when L1 based specifications fail would follow from the grammatical licensing requirement in parsing. This is the Parser as Language Acquisition Device (PLAD) as a theory of transitions for the Full Transfer/Full Access model (Schwartz & Sprouse, 1994). Changes in grammatical states as a feature reassembly problem (Lardiere, 2009) will be discussed according to the PLAD hypothesis. Hence, certain L2 developmental stages, well-known in the literature, seem in line with immediate/economy-driven parser actions. They are odd otherwise, it is argued. The problems of intuitions and of extended periods of stability with rapid changes will be discussed on the PLAD hypothesis, given general properties of the processing system. It will be argued that a parser seems to be a necessary component in the processing of L2 input sentences: a necessary prerequisite for PLAD. Various consequences and caveats are examined.
In category: Second language acquisition
Harnessing the power of words to understand science
Time: 03:00pm - 04:30pm
Place: Wells Library LI030
Staša MilojevićCognitive studies of science focus on science as a body of knowledge, i.e., ideas and relationships between ideas. Given the importance of textual documents in the practice of science it is natural to focus on the shared conceptual systems of scientific communities as expressed through the terminology used in those documents. Staša will describe some of the methods she has developed for the quantitative history of concept formation and development. She has been focusing on analyzing words derived from document titles and full text as a promising approach to examine the standardization of field vocabulary, diffusion of concepts, and cognitive extent of different scientific fields and disciplines.
JEvents v3.0.9 Stable
Copyright © 2006-2013
Powered by Joomla!®
Copyright © 2008 The Trustees of Indiana University | Copyright Complaints