Designing and Assessing an Intelligent E-Tool for Deaf Children Rosella Gennari Free University of Bozen-Bolzano Piazza Domenicani 3 - 39100 Bolzano, Italy gennari@inf.unibz.it Ornella Mich Free University of Bozen-Bolzano Piazza Domenicani 3- 39100 Bolzano, Italy mich@fbk.eu ABSTRACT LODE is a logic-based web tool for Italian deaf children who have problems in the comprehension of narratives in a verbal language; namely, it aims at stimulating global reasoning on written e-stories. Presently, LODE deals with global temporal reasoning; temporal reasoning problems are encoded as constraint satisfaction problems that can be solved by a constraint reasoner. The focus of this paper is the intelligent user interface of LODE; after characterizing the intended end users of LODE, this paper focuses on critical issues faced in the design and assessment of the interface of our e-tool. A preliminary assessment and an evaluation plan of LODE conclude our paper. Author Keywords Interfaces that teach and provide feedback, AI techniques in interfaces, intelligent visualization. ACM Classification Keywords H5.m. Information interfaces and presentation (e.g., HCI): Miscellaneous. INTRODUCTION Linguistics education of deaf children may follow the oralist method, based on the use of spoken languages, or the manual method, based on the use of sign languages. In recent years, much research has been devoted to applications for sign languages. See for example [8,13]. Less attention seems to be paid to e-learning tools for improving the literacy of deaf children in verbal languages. Our LOgic-based e-tool for DEaf children (LODE) belongs in this latter class. According to the literature [3] and our own experience, deaf people tend to have problems in formulating global relations between events of a written narrative in a verbal language. LODE aims at stimulating deaf children to globally reason on narratives written in Italian. Thus LODE presents children with e-stories and apt exercises that stimulate them to analyze the temporal relations between events, even distant in the story, and to produce new relations consistent with the story. The consistency of the story will be determined by the automated reasoner of LODE, namely, a constraint programming system. This paper, firstly, describes the LODE users; this is meant to motivate our choices in the design of the interface of LODE and present the issues encountered in the assessment of such an e-tool with deaf children. Secondly, it outlines the essential background on automated temporal reasoning and the planned architecture of our e-tool. Then it details critical aspects in the design of the user interface of LODE. A preliminary assessment and a long-term evaluation plan of LODE conclude this paper. THE END USER OF LODE The Educational Goal of LODE Learning to read and write effectively can be a difficult task for deaf people: “Deaf children have unique communication needs: unable to hear the continuous, repeated flow of language interchange around them, they are not automatically exposed to the enormous amounts of language stimulation experienced by hearing children” [20]. Research indicates that deaf people tend to reason on concept details, on single episodes of a narrative in a verbal language and show difficulties in formulating coherent global relations, such as temporal relations, between episodes of the narrative [3] to the effect that their reading ability does not often go beyond that of an eight-year old child [18]. As Arfé et al. sustain in [3], such an attitude may well depend on the kind of “literacy interventions” addressed to deaf children which tend to “focus on single sentences and the grammatical aspects of text production”. A novel literacy e-tool for deaf children should thus focus on stimulating global deductive reasoning on entire narratives. This is the main educational goal of LODE. The User Interface Clearly, in the design of the user interface of e-tools usable by deaf people, the visual input should always augment or replace the auditory input [4,21]. Moreover, captions must be provided with all multimedia presentations and all visual cues must be noticeable even if the user is not looking Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. IUI'08, January 13-16, 2008, Maspalomas, Gran Canaria, Spain. Copyright 2008 ACM 978-1-59593-987-6/ 08/ 0001 $5.00 325