O n f e r e n c e

Short paper 
Introduction 
Computer-assisted language learning (CALL) provides an efficient approach to instruction 
and practice for understanding language 
rules and assessing students’ understanding 
through computerized tests. However, when the computer program merely provides 
rules, explanations and practices, it remains the conventions of grammar teaching. Most 
online writing programs that provide stages of writing processes remain a modality of 
assessment that provides guided instruction and immediate feedback. To date, there 
exist no clear theoretical principles of second language (L2) learning to design an online 
grammar learning program that addresses the challenges for L2 learners to produce 
error-free writing. By examining relevant theories that account for grammar learning and 
second language acquisition, the study is to find and justify the theoretical principles for 
guiding the design of online grammar learning program for linguistic accuracy with 
relative automaticity. 

---

-83- 
2014 CALL Conference 
LINGUAPOLIS
www.antwerpcall.be 
Differences between grammar learning and accurate writing 
Many researchers since Krashen’s (1985) Input Hypothesis have given their attention to 
understanding how to generate effective influence of the initial L2 input on L2 output. For 
investigati
ng L2 output, the Ellis’ model shown in Figure 1 (1994, p. 349) raise the issue 
of achieving implicit knowledge. 
Figure 1. A framework for investigating L2 acquisition (Ellis, 1994, p. 349) 
How implicit knowledge differs from explicit knowledge can be 
depicted by Anderson’s 
ACT model (1976, 1983), as shown in Figure 2. Accordingly, knowledge starts out as 
“declarative knowledge” (explicit knowledge) and later transforms to the final stage of 
“procedural knowledge” (implicit knowledge). Procedural knowl
edge that people can only 
manifest in their performance differs from declarative knowledge that people can 
describe, as language acquisition is different from language learning. Procedural 
knowledge refers to using the grammar to write accurately whereas declarative 
knowledge refers to rules for grammatical accuracy. The intake of grammar in production 
memory requires procedures like matching of phonemes and recalling of appropriate 
forms. When operating efficiently, the speed of the procedures is greatly increased and 
thus frees up the workspace for further processing like composing. The computer 
program of grammatical attention and linguistic acquisition (GALA) designed in this study 
aims to assist L2 learners to develop their grammatical knowledge in the production 
memory where the knowledge can be transformed to the procedural one for execution 
needed in writing. 
Figure 2: Conceptual model of ACT