The ICICLE Project: An Intelligent Written English Tutoring System for Deaf Students

Method:

The target learner group for ICICLE is native or near-native users of American Sign Language (ASL). This population poses unique challenges for a writing instruction program: their writing contains many errors which are not made by native users of English, and students vary widely across levels of language ability, with some college-age writers producing near-native-like English and others struggling with grammatical basics. Because of these characteristics of the learner population, it is integral to ICICLE's goal of user-tailored instruction that it account for user differences so that the instruction it provides is appropriate for a learner at any level. Since ASL is a distinct and vastly different language from English (Michaud and McCoy, 1998), we view the acquisition of written English skills to be a task in second language acquisition for these learners. We are therefore designing a user model which incorporates a representation of the language acquisition process, and we have based our design upon current research in language acquisition and in the acquisition of cognitive skills.

The error identification module will use this model to determine between multiple interpretations of a sentence which may correspond to different perceived errors in the text. Some of these parses represent different structural representations of the text, and in the case of ungrammaticality may place the ''blame'' for the error on different constituents. Since determining the nature and cause of student errors is an integral step to deciding how to approach student instruction (Matz, 1982), the parser must be able to make principled decisions between these options. To determine which of these possibilities is correct, it is necessary for the error analysis component to have at its disposal a model of the student's grammatical proficiency which indicates his or her mastery of the language rules involved. This knowledge would aid in choosing between structurally-differentiated parses by providing information on which grammatical constructs the user can be expected to use correctly or incorrectly (McCoy, Pennington, and Suri, 1996).

We also wish for ICICLE to give instruction only on those language structures which are at the user's current level of acquisition; errors on structures above this level are likely to be beyond the user's understanding, while errors on structures which are well-established are likely to be simple mistakes which do not require instruction. The user model will therefore be consulted at the point where the error identification module passes the list of errors to the response generation module, trimming off those errors outside the current level.

The user model and other sources of information will also play a role in the construction of the system response. In order to structure explanations of a given language structure, the text planner needs to know the user's depth of related knowledge, including whether or not the user knows the concepts which may be mentioned in the explanation. In the final stages of response generation, the surface generator will consult the model of acquisition in order to determine which grammatical constructs are known and thus understandable to the user, and which he or she may obtain the most benefit from viewing as positive examples, in this way providing the learner with the positive input he or she needs.