The Naive Lexical Hypothesis: Evidence from Computer‐Assisted Language Learning

Abstract

COMPUTER-ASSISTED LANGUAGE LEARNING(CALL) environments provide tools such as word-processing and database searching which encourage creative interaction by students with a rich environment for foreign language learning supported by databases of reference materials.! The new technology also encourages the direct involvement of instructors in the learning process, by offering the possibility of recording student responses and response paths in real time. This previously inaccessible "window" into the students' searches in the dictionary or grammar reference during the composing process now permits new insights into the way students process language information. Not only does this possibility permit evaluation of the effectiveness of the instructional software and setting, but it also raises interesting questions about the teaching and learning processes. How do learners interact with this new learning environment? How do they exploit the computerized resources? What does studentcomputer interaction imply about learning and teaching?We sought to answer these questions in a pilot study of first-and second-year learners of French attempting composition assignments with a fully supported and documented software system for foreign language learning called Systeme-D (21).2 Systeme-D is a writing tool for learners of French which combines a bilingual word-processor with an interconnected set oflanguage reference materials consisting of a bilingual dictionary, a reference grammar with usage notes, and a thematic collection of sit-The Modern Language Journal, 74, iv (1990) 0026-7902/90/0004/440 $1. 50/0 ©1990 The Modern Language Journal uational/topicallfunctional vocabulary and phrases. We tracked the learner's ongoing interaction with these resources during a composing task by recording keystrokes on the microcomputer; combinations of keystrokes were designated as particular "query events," which could then be related to changes and additions in the student composition. This paper reports on the use of Systeme-D as a research tool to investigate the types of queries made by the language learner during the composing process. These queries represent different strategies for trying to construct meaning in L2, and they suggest different types of developmental interlanguage hypotheses made by classroom L2 learners.One of the most pervasive features of the linguistic behavior of the beginning language learner is the attempt at direct translation. The initial assumption by the learner seems to be that for every word in L1 there is a one-to-one lexical match in L2. We have called this assumption or heuristic about L2 the naive lexical hypothesis because of the lexical (rather than grammatical) focus of the learner's thinking. The purpose of this paper is to examine the naive lexical hypothesis (NLH) as it is revealed in student query behavior at the computer. We will explore the relationship of the NLH to three different classes of learner queries which represent different levels of sophistication in interlanguage hypotheses. BAC...

How do we Know what Students are Actually Doing? Monitoring Students' Behavior in CALL†

“…Similarly, while using ESL/EFL reading courseware with multiple types of annotations, Laufer and Hill (2000) found that students predominantly consulted information on word meanings. In a pair of classic studies, working with Système-D (a French writing assistant program for students that contained a bilingual dictionary, a functional reference grammar, and a verb conjugator), Bland, Noblitt, Gay, and Armington (1990) and Noblitt and Bland (1991) remarked that students used the dictionary for English-to-French translations fully 80% of the time. Hulstijn (1993) found great variability in students' consultations of word definitions in a CALL-based reading program and observed that some students skipped entire paragraphs in the reading passage even though they knew they were going to be tested on the content of the passage (also see Hwu, 2004).…”

Coherence and cohesion: An analysis of the final project abstracts of the undergraduate students of PGRI Semarang

“…That is why the difficulty in mastering these language components, grammar and vocabulary will of course contribute to the inability to write a text well. As cited by Kern (2000: 171) from the studies conducted by Bland, Noblitt, Armington and Gay (1990), the early learners of a new language were dominated by the difficulty in vocabulary; while the advanced learners as students in universities are dominated by the genres and rhetorical organization problems. Consequently, the text they write often does not show its unity or incoherent for they don't know much how to link This problem leads the writer to investigate whether or not the students' abstracts of their final project reports are coherent and cohesive.…”

A quantitative interaction analysis of an authentic CALL environment: New Dynamic English

“…The data gathered from Research Question One corroborate earlier findings by Hsu, Chapelle, and Thompson (1993) and Bland, et al (1990), which show that learners do not access all the available information when working in an interactive CALL environment. As mentioned earlier, even though the repeat button was the only option for modification of input that was explicitly introduced to the students, 15 of the 91 still did not use it.…”

“…The following were considered options for the user: the ABC button, the repeat button, the glossary, the microphone button, the headphone button, and a speech recognition (SR) task attempt. Hypothesis 1 (HI) was constructed based upon past research by Bland, et al (1990) and Hsu, Chapelle, and Thompson (1993): this research was expected to reveal that students would use some options more than other options and that this choice would vary from student to student.…”