Skip to main content

IA CENTRADA EN LAS PERSONAS: INTERFACES ADAPTATIVAS DIRIGIDAS POR EL USUARIO

PID2023-148577OB-C22

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Call year 2023
State Plan 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSIDAD COMPLUTENSE DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Beyond the Blank Prompt: Co-Designing Accessible Generative AI Interfaces for People with Intellectual Disabilities

  • Francisco Gilmartín, Virginia
  • Hervás Ballesteros, Raquel
  • Chinchay, Yussy
Generative Artificial Intelligence (GenAI) systems offer new opportunities for access to information, content creation, and decision-making, but their dominant interaction paradigm poses important barriers for people with intellectual disabilities (ID). In particular, open-ended prompts, long and poorly structured responses, and limited support for refinement, verification, and personalization can increase cognitive load and hinder effective use. This paper presents an exploratory, expert-informed co-design workshop aimed at rethinking GenAI interfaces from the perspective of cognitive accessibility for people with ID. The workshop involved ten researchers with expertise in accessibility, Human–Computer Interaction, Artificial Intelligence, and inclusive design, and was structured around five key interaction challenges identified in previous empirical work: question generation, response refinement, information verification, context management, and personalization. The workshop generated a set of preliminary design concepts that point toward more guided, structured, and adaptive interaction models, including mechanisms such as step-by-step prompting, explicit refinement controls, support for source interpretation, conversation structuring, and dynamic personalization. Rather than presenting validated interface solutions, this work offers an initial map of design opportunities for cognitively accessible GenAI and provides a foundation for future participatory design, prototyping, and evaluation with people with ID.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Thinking Outside the [Chat]Box: Bridging Computer Science and Industrial Design for Cognitive-Inclusive Generative AI

  • Francisco Gilmartín, Virginia
  • Guasch, Daniel
  • Hervás Ballesteros, Raquel
Current Generative AI (GenAI) interfaces remain largely constrained to chatbox interaction, which can impose high cognitive demands on users and create substantial barriers for people with intellectual disabilities (ID), including prompt formulation difficulties, response overload, and limited mechanisms to assess information reliability. To explore alternative interaction models for cognitive accessibility, we conducted a cross-disciplinary co-design challenge in which two student cohorts (Computer Science and Industrial Design) developed interface concepts from the same set of functional requirements (e.g., prompt scaffolding, structured output, GUI-based refinement, transparency, and personalization). Comparing the resulting proposals reveals both convergence on foundational requirements (notably initial calibration, proactive prompting, and direct manipulation of response fragments) and complementary contributions that outline a multi-layered support system. Computer Science teams primarily produced structural scaffolding, emphasizing predictability, navigability, and trust through mechanisms such as reliability indicators, explicit sources, and context management for long conversations. Industrial Design teams emphasized experiential scaffolding, focusing on pacing, attention guidance, multimodality, and proactive agency, including step-by-step response flows, focus modes, and assistant-like integrations. We synthesize these findings into a dual-layer scaffolding framework that expands the design space for cognitively accessible GenAI interaction beyond chat-centric models and motivates future work on expert refinement, technical feasibility, and empirical validation with users with ID.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Challenges in accessing generative AI for users with cognitive disabilities: an exploratory case study

  • Francisco Gilmartín, Virginia
  • Hervás Ballesteros, Raquel
  • García-Mata, Ricardo
Generative Artificial Intelligences (GAIs) have great potential for people with cognitive disabilities, since they can increase their autonomy, offer them personalized learning and break down the barriers that this group traditionally has to access information. To address this gap, this paper investigates the specific difficulties experienced by users with cognitive disabilities when interacting with GAIs. Wepresent an exploratory case study involving 16 students with cognitive disabilities enrolled in a universitybased educational program. ChatGPT was selected as the representative GAI tool due to its popularity and the similarity of its interface and interaction paradigms to those of other widely used Generative AI systems. The intervention was integrated into the students’ academic curriculum and involved completing practical academic tasks using ChatGPT. A mixed methodological approach was employed, combining pre- and post-intervention questionnaires completed by students and tutors/experts, along with on-site observations during the sessions. Results of this exploratory study corroborate the existence of remarkable barriers in basic interaction areas: question formulation (69% of students required support), response comprehension (56% reported understanding answers only occasionally), and information verification (only 19% were able to verify the system responses independently). Suggestions for improvement, from the perspective of the users and their tutor, include more guided interfaces with predefined question templates and contextual suggestions, customizable output formats with language simplification, enhanced spelling correction tools, interactive explanations of complex terms and built-in verification aids. This study highlights the need to design GAI interfaces with a strong focus on cognitive accessibility. Future development should prioritize multimodal and adaptive interaction models, support tools for comprehension and verification, and interaction designs that promote greater autonomy for users with cognitive disabilities.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Combining the user friendliness of SignWriting with the precision of linguistic parameters

  • García Sevilla, Antonio Fernando
  • Lahoz Bengoechea, José María
  • Conde González, Sandra
  • Díaz Esteban, Alberto
  • Folgueira Galán, Pablo
  • Calle Pérez, Julia de la
This paper presents TraduSE, a Progressive Web Application designed to bridge the gap between the user-friendliness of SignWriting and the precision required for sign language processing tasks. Capturing sign language complexity is crucial, but parameter-based approaches often have a steep learning curve and require specialized tools. SignWriting offers a more accessible alternative for representing signs, and TraduSE addresses the challenge of making linguistically well-informed systems accessible to signers, learners, and researchers by using SignWriting as an input for a native sign language dictionary: the Spanish Sign Language Signary.It does so by allowing users to upload or draw SignWriting images, which are recognized by an existing artificial vision pipeline. The SignWriting elements found are then converted into a parametric description, recovering along the way the 3D signing space information from the 2D SignWriting. Finally, the resulting parameters are used to search the Spanish Sign Language Signary. Fuzzy matching allows for error resilience and finding similar tokens, enabling users to locate signs even with incomplete or imprecise information. Search results are presented as videos and Spanish glosses, while also displaying and explaining the linguistic parameters, serving as a didactic tool.Our approach not only provides a crucial bridge between visual sign representation and precise linguistic analysis, widening the reach of sign language research, but also demonstrates flexibility in handling the inherent fuzziness and potential errors in everyday language use and user input.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Translating Easy-to-Read Standards into Prompts: An Empirical Study of LLM-Based Text Simplification Strategies

  • Broto Ortega, Juan
  • Francisco Gilmartín, Virginia
  • Hervás Ballesteros, Raquel
Text simplification remains a significant challenge, especially when texts must comply with Easy-to-Read (E2R) standards. This paper explores how normative E2R guidelines can be translated into prompt configurations for Large Language Models (LLMs) to support text simplification. We compare six prompting strategies, ranging from a baseline prompt without explicitly including the UNE guidelines to one-shot and few-shot strategies. Using Gemma 3 4B, we evaluate the generated simplifications on a mixed corpus of human-written and synthetic texts, through automatic linguistic complexity metrics, and human evaluation of guideline compliance and content preservation. Results indicate that prompting strategies that explicitly incorporate the guidelines, particularly those including examples, improve compliance with UNE-based E2R requirements compared to baseline configurations. Overall, this study provides initial evidence of the potential of LLMs as a support tool for generating E2R adapted content, and also highlights the importance of future evaluation with end users.