Content

This page presents an overview of this course on ICALL as the following topics will be covered. The items in the list are ordered according to the planned schedule and not according to the table of contents.

1. In the first part the view of the learner is presented combined with the views taken in current didactics on computers and language teaching. It will be shown, that the dramatic increase in computing and communication power has led to an increased usage of computers in language teaching besides other applications. New types of exercises seem to have been established in language teaching with the aid of computers which enable the learner to access a foreign language with (possibly) new methods.
   Also the trends in language didactics seem to go along well with the development of information technology (IT): For example one of the goals is to allow the learner to use so called explorative learning. With the help of the WWW and some well structured exercises, this can be achieved in a straight forward way.
2. Then a short part about the history of CALL and more important ICALL is presented. This part should enable the reader to put the following chapters into a "developmental" context. The history demonstrates how closely interwoven the development of computer technology and of the usage of computers in language classes is. Some authors have even asked, whether the development of teaching might be driven by technology. More specifically the history of ICALL can give an impression of where researchers have seen and still see interesting areas to apply methods of computational linguistics to improve CALL-software or to implement completely new types of programs. First and foremost this is of course applying natural language analysis to learner input, but also other options are available as will be seen in chapter 3.
3. Following this, two (historical) example systems are presented, which show the main direction of research development in the last decades of the last century, namely to implement dialog oriented language tutoring systems. An important aspect here is (of course) the difficulties the developers had in developing the systems in the area of computational linguistics and, less important, in other fields. In both cases almost complete dialog-systems were implemented in order to be able to analyse learner input and to generate appropriate output. These systems therefore allowed the learner to (almost) freely produce language in order to enhance her/her so called communicative competence.
4. In order to achieve a further insight into the area of actual ICALL-systems, a categorization scheme can be employed. This is especially important with the language learner/teacher in mind, because a program should be used at the right level at the right stage of the teaching process. Furthermore especially ICALL-systems, which are usually only prototypes and not complete self-learn-programs, are developed to be used in certain learning situations.
5. One important aspect about developing a program is an evaluation with respect to the intended usage. There are several aspects, which must be considered when evaluating a program for language learning. A certain evaluation scheme must be changed additionally, if ICALL-systems instead of traditional CALL-programs are to be evaluated. Considering an evaluation scheme also helps in the development of a new ICALL-system. Since the technical aspects of a project are usually in the foreground, an evaluation scheme might help also to keep the learner and his needs in mind.
6. The chapter on functionality tries to show the variety of places, where an integration of CL-techniques and linguistic resources can and has been tried. This ranges from superficially simple orthographic checking to syntax error recognition to linguistic databases. Basically two main areas can be distinguished: Methods (or tools) and resources. "Methods" in this context means applications, which do either analysis or generation of language. Resources however are not used by an application per se, but are usually data bases with linguistic information, which can be retrieved for presentation to the learner.
7. The following chapter deals with error analysis, and more specifically with the recognition of syntactic errors in learner language. This chapter is taken out of the list of error analysis subchapters, because in the field of CL this aspect has received the most attention. In general one can say, that there are two ends on how to recognize syntactic errors with NLP-tools. Either a parser is developed, in which the parsing algorithm is capable of identifying errors or the linguistic information, i.e. the grammar and/or the lexicon contain information about possible errors. Additionally a mixed approach can be chosen in order to exploit the differences in the predictability of errors.
8. Following the treatment of syntactic errors a look is taken at the recognition of errors in other linguistic levels, such as phonology, morphology and semantics. These fields are dealt with together, because only little reseach has been carried out here. Nevertheless these fields are important areas in language teaching and therefore should be mentioned.
9. The topic of this part is the construction of intelligent tutoring systems (^↗ITS). As the name suggests these systems are designed to tutor a user with respect to certain learning material. Here the chapter tries to abtract away from the content and offers insight into the advantages and difficulties related to ITSs in general. There are several modules which are suggested to manage the interpretation of the learner input, the inference with respect to the internal model, the presentation of pedagogically sound follow-up actions etc.
10. The last item of the course deals with learner modelling in ICALL-systems. Here the inferencing is not in the foreground but the general possibilities for adaptation of the system to the language learner. There are various ways of adaptation with respect to the content, to feedback, to help-messages etc. On the one hand these adaptive measures should adapt the system to the user and on the other hand this should enable the system to provide the right learning material and exercises at the right level of progression, i.e. challenge the learner.