SLATE LECTURE SERIES

2000-2001

SLATE is pleased to announce the speakers for this semester. All times are 7:30-8:30 with reception following. Stay tuned for more information.


November 29, 2000

Peter Golato, Professor of French, UIUC

Title: Operationalizing Language Dominance in Late-learning French-English Bilinguals
Place: Lucy Ellis Lounge (Foreign Languages Building)
Time: Wednesday, November 29, 2000, 7:30 pm

In operationalizing language dominance in bilinguals, previous studies have used various forms of self-evaluation. While such categorical measures may reveal which of a non-balanced bilingual's languages is dominant, it remains to be seen how a self-evaluation of language dominance is related to actual language perception and production skills. In this talk, I will present a scalar measure of language dominance which is based upon language perception and production data, a measure which potentially offers a finer-grained portrait of language dominance than does self-evaluation.


October 12, 2000

Bill VanPatten, Professor of Spanish and SLA (University of Illinois at Chicago)

Title: A Brief History of Input in SLA and Why we Need to Look at Processing
Place: Lucy Ellis Lounge, FLB
Time: 7:30 p.m.

In this talk, I will trace the construct of input in SLA theory and how the construct has evolved since Corder's seminal publication in 1967. I will also review several mainstream theories in SLA to explicate the role of input in those theories. What I will argue is that in spite of our advances in the field of SLA theory, little is known about how learners interact with input to create the data necessary for language acquisition. I will then outline a model of input processing and discuss what the field gains by focusing its attention on how learners process input.

September 18, 2000

Irene Koshik (University of Illinois at Urbana-Champaign)

Title: Categorization in SLA research on L2 discourse: The importance of attending to courses of action.
Place: Lucy Ellis Lounge, FLB
Time: 7:30 p.m.

This talk will show how conversation analysis methodology, which analyzes courses of action implemented through talk, can provide new insights into aspects of teacher/student talk which are unavailable when utterances are taken from their sequential context and coded according to pre-determined categories. Mehan's (1979, 1985) initiation/reply/evaluation (IRE) sequence structure, initially used to characterize a specific speech event, i.e., recitation sessions in elementary school classrooms, has often been assumed to describe pedagogical talk in general, and has been used to categorize a variety of different pedagogical speech events. Teacher questions have been coded according to whether they ask for students to display knowledge which the teacher already knows, as in the initiation turns of the IRE sequence, or whether they are asking for unknown information, as in conversation. In the L2 pedagogical literature, the two question types are referred to respectively as "display" and "referential" questions (e.g. Long & Sato 1983, Brock 1986, Pica & Long 1986, Chaudron 1988, Tollefson 1988, Allwright & Bailey 1991, Markee 1995). I will provide an analysis of three sequences of teacher/student talk in the same type of speech event: a one-one writing conference. Each sequence contains teacher questions which could be categorized as "display" questions. Although the sequences contain elements of the IRE pattern, I will show that each sequence has a more complex structure which sustains a particular course of action and assists the student to understand and participate in that course of action. Teachers' "known-answer" questions are not used merely to elicit knowledge displays but are used for a variety of purposes within these courses of action. Categorizing these questions prematurely, without a close analysis of the talk in its sequential context, can prevent us from seeing what these questions accomplish pedagogically. Similarly, categorizing pedagogical discourse into a simple IRE pattern can oversimplify the structure of the sequences in which known answers are elicited and can limit our understanding of how sequence design contributes to the action which the turn is being used to accomplish.

 

Thursday, March 29, 2001

Carol A. Chapelle, Iowa State University

Title: Evaluating CALL: Questions for the 21st Century
Place: Lucy Ellis Lounge (Foreign Languages Building)
Time: Thursday, March 29, 2001, 7:30 pm

In an era when many language learners and teachers assume that technology should be one vehicle for second language learning, most researchers agree that investigations of computer-assisted language learning (CALL) are too limited if they seek only to compare learning outcomes of CALL users with those of learners in traditional classrooms. Despite the definitive move away from such comparisons as a sole research strategy, the intuitive appeal in demonstrating superior effects from technology has kept this research tradition alive. Moreover, it would be difficult to argue that the results of such research are uninteresting. What is the problem with the control-treatment study for investigating CALL? What alternative approaches have been attempted, and how can the various research approaches be seen as complimentary sources of information? This paper will explain an approach to identifying important questions about CALL and applying relevant research methods to address these questions. It will suggest that the issue for the 21st century is how best to integrate multiple sources of research results to evaluate CALL.


Thursday, April 5, 2001

Michael T. Ullman (Departments of Neuroscience and Linguistics, Georgetown University)

Title: The Declarative/Procedural Model of Language: Extensions to Sex Differences and Second Language
Place: Lucy Ellis Lounge (Foreign Languages Building)
Time: Thursday, April 5, 2001, 7:30 pm

Our use of language depends upon two capacities: a mental lexicon of memorized words, and a mental grammar of rules that underlie the productive sequential and hierarchical composition of lexical forms into complex linguistic representations--i.e., complex words, phrases and sentences. The Declarative/Procedural model posits that the learning and use of lexical knowledge depends upon a well-studied bilateral temporal-lobe "declarative memory" system implicated in the learning and use of conceptual/semantic knowledge (i.e., knowledge about the world), while grammatical computations that underlie the real-time combination of lexical forms into complex representations rely on left frontal/basal-ganglia "procedural" circuits implicated in the acquisition and expression of motor and cognitive skills (e.g., riding a bicycle).

The Declarative/Procedural model predicts double dissociations between lexicon and grammar, with associations among lexical memory, memorized facts, and temporal-lobe structures, and among grammar, motor skills, and frontal/basal-ganglia structures. The model is supported by studies investigating morphology and syntax; using a range of psycholinguistic and neurolinguistic approaches, including behavioral testing of patients with aphasia, neurodegenerative disease or developmental disorders, and neuroimaging investigations of healthy subjects (using fMRI, MEG, and EEG/ERP); with children and adults; examining several languages (English, German, Japanese, and Italian).

Two extensions of the model are discussed. First, sex differences in the neurocognition of lexicon and grammar are examined. Robust evidence indicates that females are better than males at remembering words. This suggests the novel hypothesis that females may tend to memorize previously-encountered complex forms (e.g., played), while males generally compute these forms compositionally (e.g., play + -ed). Both sexes should compute new complex forms compositionally (e.g., proy + -ed). These predictions are confirmed with converging evidence from psycholinguistic, neuropsychological, and neuro-electrophysiological studies examining the processing of complex words and sentences.

Second, neurocognitive differences between first and second language are examined. Evidence suggests a critical (sensitive) period in the acquisition and use of grammar: Older learners have greater difficulty than younger learners. This leads to the hypothesis that older second language learners, being unable to depend upon the procedural/grammatical system, are forced to rely on the declarative/lexical system for the computation of complex linguistic representations. These representations may be either memorized, or constructed by explicit rules learned in declarative memory. This shift to declarative/lexical memory is expected to increase with increasing age of exposure to the language, and with less experience (practice) with the language, which is predicted to improve the procedural/grammatical learning of grammatical rules. Evidence is presented in support of these predictions.

See Spring 2000 SLATE Lectures
See Fall 1999 SLATE Lectures