Collaborative design (or co-design) involving users with disabilities plays a critical role in the development of accessible and user-centred digital solutions. However, most existing applications for accessible text adaptation are created without the active involvement of the people they aim to support. To address this gap, this work presents a participatory design process carried out with individuals with cognitive disabilities to collaboratively shape the user interface of AlicIA, a web- and mobile-based application for adapting short stories in Spanish into accessible versions, following the Easy-to-Read (E2R) Methodology. The feedback gathered during the co-design session served as a basis for concrete decisions on the design of the user interface, the inclusion of accessibility features, and the use of visual aids such as pictograms. Framed within the principles of Responsible Artificial Intelligence (AI), this work underscores the importance of recognising users with cognitive disabilities as active contributors in the development of AI-supported tools for cognitive accessibility.
Access to information is a fundamental right and a key condition for full social participation. However, there are diferent groups of people who present dificulties in comprehension, including people with cognitive or intellectual disabilities. Since these groups have the right1 to actively participate in aspects of society, such as politics, education, work, and culture, on an equal basis with others, achieving cognitive accessibility is crucial to enabling inclusive communication and understanding.
Culture, in particular, plays an essential role in transmitting knowledge, fostering identity, and
encouraging self-reflection. For people with cognitive disabilities, being able to engage in leisure
activities, especially through the appreciation and understanding of literary texts, is not only enriching
but also a matter of inclusion2. To support this right, cultural materials, and literary content in particular,
must be made accessible in formats adapted to diverse comprehension needs. In this sense, the
Easy-toRead (E2R) Methodology [
At the same time, technology plays a central role in daily life, facilitating communication and access to information. However, digital environments are not always inclusive. In particular, people with disabilities frequently encounter significant barriers when using digital platforms. This has led CEUR
ceur-ws.org
to important initiatives such as the Web Content Accessibility Guidelines (WCAG) [
This work focusses on the inclusiveness dimension of Responsible Artificial Intelligence (AI),
understood as the imperative to ensure that individuals with disabilities are not excluded from AI systems,
but are instead actively involved in their design and development [
The rest of the paper is organised as follows: in Section 2, we present the State of the Art, including an overview of co-design methodologies with people with cognitive disabilities and the existing tools for accessible text adaptation. Section 3 describes the participatory design process conducted with users with cognitive disabilities. In Section 4 we introduce how the co-design process results were used in the implementation of AlicIA user interface. Finally, Section 5 presents some conclusions and future work.
In this section, we provide (a) an overview of collaborative design and its relationship with cognitive disability, and (b) a review of current technological tools aimed at making texts more accessible.
Collaborative design (or co-design) is increasingly recognised as a key methodology for developing
inclusive technologies, particularly when the target audience includes people with cognitive or
intellectual disabilities [
However, engaging people with cognitive disabilities in design processes presents unique challenges.
Traditional co-design methods often rely on verbal expression, abstract reasoning, or symbolic thinking,
which are skills that may be limited in this population [
Recent work has also highlighted the importance of relational and embodied forms of participation,
where communication may be expressed through non-verbal means, afective responses, or subtle
behavioural cues [
Numerous web applications and tools have emerged in recent years to support text accessibility and simplification, especially for users with cognitive disabilities.
Regarding technological support to address the E2R guidelines and recommendations in Spanish texts,
it is worth mentioning (a) Easy-to-Read Advisor [
In addition to these, Clara3 ofers a data-driven approach to clarity evaluation. Instead of simplifying text, it predicts the comprehensibility of documents based on a set of nine linguistic and structural metrics, providing objective insights into accessibility. LeeFácil [18], for instance, is a prototype that extracts text from images and provides lexical support through definitions, synonyms, and pictograms. Although it does not simplify full texts, it contributes to enhancing lexical comprehension. EasyReading [19], a browser extension for Chrome and Firefox, includes features such as text-to-speech, layout adjustment, and pictogram overlays. Although its easy-to-read rewriting module was not operational during testing, its visual aids help support users with cognitive disabilities. PLACEAT4 provides three types of adaptations: a simplified text, a pictogram-enhanced version, and a list of comprehension questions, along with a glossary. However, the application has not been updated since 2022.
These tools reflect the increasing interest in automating and supporting the creation of accessible content. Nevertheless, while some recent systems such as EASIER have adopted user-centred approaches and carried out evaluations with users with cognitive disabilities [20], most existing applications have been developed without involving users in the design of their user interfaces. In particular, participatory design sessions, where users are involved as co-creators rather than just testers, are still not adopted, especially in applications for adapting narrative literary texts such as short stories. To cover these gaps, we developed AlicIA, a tool for adapting short stories in Spanish into easier versions. A user-centred design process that included a participatory design session with individuals with cognitive disabilities was applied for designing the user interface of AlicIA.
As mentioned in Section 1, this work is framed within the principles of Responsible Artificial Intelligence (AI), with a particular emphasis on inclusiveness. In this context, we adopted a human-centred approach [21] to ensure that the application reflects the needs and preferences of users with cognitive disabilities. According to Trewin [21], there are three main approaches to incorporating people with cognitive disabilities into the development of digital technologies: Inclusive Design, Participatory Design, and Value-Sensitive Design. In our case, we opted for Participatory Design, which actively involves end users in the design process and treats them as co-creators rather than passive recipients [22].
We organised a participatory design session with people with cognitive disabilities to establish the basis for the development of the AlicIA user interface. The session took place on April 30 2024, and lasted approximately one hour. The participants included adults with cognitive disabilities, who were also E2R validators, from the ACCEDES organisation5, based in Madrid (Spain). The participatory design session was structured following the methodological principles of a focus group, a qualitative technique particularly useful for eliciting shared knowledge and collective reflection among participants [23]. Focus groups are especially valuable for understanding not only individual opinions but also how participants interact, agree, or disagree around a given topic [24]. The session was moderated by the principal investigators of AlicIA, together with the support professional who regularly works with the participants. This professional played a key role in interpreting the mock-up proposals, rephrasing the questions when needed, and encouraging the participants to share their preferences. As noted by Lane et al. [23], the role of the moderator in focus groups is crucial to fostering inclusive participation, particularly through active listening and relational sensitivity.
To facilitate discussion, we prepared a PowerPoint presentation containing a series of mock-ups6, 3https://clara.comunicacionclara.com/ 4https://placeat.org/asistente-de-lectura-facil-placeat/ 5https://accedes.es/ 6Available at: https://zenodo.org/records/15682729 (In Spanish) each illustrating diferent design alternatives for the AlicIA user interface. Although AlicIA is also a mobile application, it is important to note that all the mock-ups and feedback discussed in this session were focused on the design of the web application. The mock-ups were structured to explore various aspects of the application, including: a Text input area: Participants were presented with two options: one showing separate text boxes for the original and adapted text, and the other one combining both functions into a single box.
Participants were asked to indicate which layout they found more intuitive to use. b Instruction wording: Participants were shown two options to choose how the input instructions should be presented. Option 1 showed the instruction Escribe tu relato debajo o escoge un archivo clearly displayed above the text input box, framed in a coloured banner. In Option 2 the instruction was integrated directly into the text box as placeholder text: Escribe o pega aquí tu relato. También puedes escoger un relato con el botón “Escoge un archivo”. c Pictogram support: A sample of adapted text was presented with corresponding pictograms.
Participants were asked whether this visual aid helped them understand the text more easily. d Text-to-Speech feature: The possibility of listening to the adapted story through a text-tospeech (TTS) function was introduced, and participants were asked for their opinions on its usefulness and preferred form of activation.
A total of 4 participants took part in the co-design session. The group consisted of 2 females and 2 males. Three participants were Spanish nationals, and one was Peruvian. In terms of age, one participant was 21, one was 33, one was 42, and one was aged 43. Regarding disability profiles, two participants had intellectual disabilities, one had a neurological condition, and one had both an intellectual disability and a mental health condition. Reading comprehension levels7 varied across the group: one participant demonstrated a low level, one a high level, and two reported a medium-high level of comprehension.
We are aware that the main limitation of this co-design session is the small sample size. However, in qualitative research, sample size is not the main concern. As Parahoo [25] points out, while group size matters during planning, it “is not a starting point” in qualitative research. In the context of focus groups, Krueger [26] and Then [27] recommend groups of 6 to 8 participants, but they also recognise that smaller groups can be valid and efective, particularly when working with vulnerable populations. For example, Lane and colleagues [23] describe sessions with only three participants due to logistical barriers, such as caregiving responsibilities or limited support.
In this section we describe the design implementations of AlicIA user interface based on the feedback provided by the participants. AlicIA is a web- and mobile-based application8 designed to support the automatic adaptation of short stories in Spanish according to the E2R guidelines, by means of Artificial Intelligence (AI)-based techniques. The application aims to promote inclusive reading practices, especially for users with intellectual or cognitive disabilities. The current version of AlicIA user interface is the result of the feedback gathered in the participatory design process described in Section 3. AlicIA is built around three main functionalities: (a) automatic adaptation of short stories into a format close to the E2R Methodology; (b) visualisation of the adapted texts with pictograms, enabling enhanced comprehension through symbol-based representations, and (c) customisable accessibility settings, including visual adjustments such as font size, letter and line spacing, and high-contrast mode. 7This information was provided by the support professional. 8The application is not yet available but will be in a few months.
As mentioned in Section 3, the participatory design session was focused on the web version of AlicIA. However, we have also developed a first prototype of the mobile app, following the same design suggestions and accessibility features shared by the participants.
The user interface reflects several suggestions made by the participants during the co-design session. In terms of visual design, the participants expressed a preference for a clean background over a gradient version shown in one of the mock-ups, which they found visually distracting. They also suggested simplifying the subtitle under the logo: the original Relatos accesibles was replaced by Aplicación para adaptar relatos. Furthermore, the welcome message was reformulated in a more friendly tone, becoming ¡Hola, soy Alicia! Te ayudo adaptar relatos, as shown in Figure 1 for the web user interface and Figure 5a for the mobile user interface.
On instruction wording, participants selected Option 2, in which the instruction was integrated directly inside the box with the instruction Escribe o pega aquí tu relato. También puedes escoger un relato con el botón “Escoge un archivo” (see Figures 1 and 5a).
Regarding the text input area, participants preferred a single text box for both the original short story and its adapted version. Additionally, participants suggested using colour-coded borders to diferentiate the two text areas: blue for the original (see Figure 2a), and green for the adapted version (see Figure 2b), since they claimed that green was intuitively associated with simplicity and ease.
With respect to buttons and user interface consistency, in the user interface view where the adapted story is displayed, participants expressed a preference for having the buttons aligned horizontally on the same level and using a unified colour scheme. Some button labels were also revised following feedback: Cargar relato was updated to Escoge un archivo, while Ver el relato con pictogramas and Adapta otro relato was retained, as participants found it clear and efective.
In addition to existing settings for font size and contrast, participants requested controls for adjusting
letter and line spacing, which is an important feature for readers with visual processing dificulties. This
was implemented following the WCAG 2.2 guidelines [
The participants strongly supported the integration of pictograms as a comprehension aid. Pictograms have already been shown to be efective in the communication of people with cognitive disabilities [29]. In the initial implementation, pictograms were displayed alongside the full adapted text. However, this layout caused visual clutter and made it dificult for users to focus on specific information. Based on the feedback of the participants, the current version of AlicIA presents one sentence at a time, with pictograms illustrating key words in sequence, as shown in Figures 4 (web user interface) and Figure 5b (mobile user interface). This structure reduces cognitive load and fosters a more focused and semantically coherent reading experience. Technically, AlicIA makes use of the ARASAAC9 API, in particular, its word-by-word search function, which accurately selects pictograms and reduces visual noise. This strategy is in line with the principles of the so-called pictotranslation10, where simplified or adapted content is enriched with symbolic representations to reinforce meaning.
One of the identified needs was to make error alerts visible to users. Previously, when an error occurred, either due to a server failure or incorrect input, there was no clear indication that something had gone wrong, which often led to confusion. To address this, we designed a set of possible error messages and asked the same group of participants, outside the co-design session, to help us. We sent them examples of diferent error situations, and they gave us suggestions on how the messages should be written. Their feedback helped us adjust the messages to make them easier to understand, as shown in Table 1. 9https://arasaac.org/ 10https://avanzandoconemociones.com/2020/07/29/crear-pictotraduccion/ Displayed Message Por un problema interno, por favor vuelve a intentarlo más tarde.
Por un problema interno, por favor vuelve a intentarlo más tarde.
AlicIA no entiende lo que escribes o no encuentra la respuesta. (a) General view.
(b) Pictogram view.
(c) Contrast mode.
In this work we present the co-design of the user interface of AlicIA, a web- and mobile-based application for the automatic adaptation of short stories in Spanish according to the Easy-to-Read Methodology guidelines. A key contribution of this work lies in the integration of a participatory co-design process, in which users with cognitive disabilities actively contributed to the design of the user interface. The participants’ feedback provided the basis for the design of the application: from visual design to accessibility features and the display of pictograms. The resulting application demonstrates how collaborative design methods can improve both usability and inclusion in AI-based accessibility technologies. We are aware that this first stage of development involved a single co-design session, focused on the web version of the application and primarily addressing interface-level aspects. Due to time and resource constraints, some components of the system were developed prior to the co-design session, and user participation was limited to a specific phase of the process. In future iterations, we aim to extend user involvement to additional stages of development, including the evaluation and refinement of underlying features, in order to ensure a more comprehensive and participatory design approach.
As future work, we will first validate the mobile version of AlicIA to see if the design feedback from the web version also works well on mobile. After that, we plan to carry out a usability evaluation of both versions, web and mobile, involving more users with cognitive disabilities. Finally, we will consider incorporating cost-of-interaction indicators and extending the reference framework with more recent literature on participatory design and cognitive accessibility.
This work has been supported by the grant “Ayudas para la contratación de personal investigador predoctoral en formación para el año 2022” funded by Comunidad Autónoma de Madrid (Spain). We would like to thank Isa Cano for her help in organising the participatory design session. In addition, we really appreciate the collaboration provided by ACCEDES (Entornos y Servicios Accesibles SL.) and its cognitive accessibility validation team, made up by persons with cognitive disabilities, from the “Así Mejor” Program of workshops and activities of the Tres Cantos City Council (Madrid).
During the preparation of this work, the authors used the AI-based tools ChatGPT and Writefull to assist with grammar and spelling checks. The authors have reviewed and edited the content as needed and assume full responsibility for the accuracy and integrity of the final manuscript. gración en la Comunidad, Universidad de Salamanca, 2019. URL: https://eapoyo-inico.usal.es/ asistente-lectura-facil/. [14] H. Saggion, S. Stajner, S. Bott, S. Mille, L. Rello, B. Drndarevic, Making It Simplext: Implementation and Evaluation of a Text Simplification System for Spanish, ACM Transactions on Accessible Computing 6 (2015). doi:10.1145/2738046. [15] S. Bott, L. Rello, B. Drndarevic, H. Saggion, Can Spanish be simpler? LexSiS: Lexical simplification for Spanish, in: M. Kay, C. Boitet (Eds.), Proceedings of COLING 2012, The COLING 2012 Organizing Committee, Mumbai, India, 2012, pp. 357–374. URL: https://aclanthology.org/C12-1023/. [16] L. Rello, R. Baeza-Yates, H. Saggion, Dyswebxia: More accessible texts for people with dyslexia [text in spanish], Procesamiento del Lenguaje Natural 51 (2013) 205–208. [17] L. Moreno, R. Alarcon, R. Martínez, Easier system. language resources for cognitive accessibility, in: Proceedings of the 22nd International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS ’20), ACM, 2020, pp. 1–3. doi:10.1145/3373625.3418006. [18] E. M. A. Selva, I. S. Soltero, Asistente móvil para la interpretación de texto dirigido a personas con discapacidad cognitiva, 2019. URL: https://hdl.handle.net/20.500.14352/15193, curso 2018/2019. [19] P. Heumader, K. Miesenberger, R. Koutny, The easyreading framework – Keep the user at the digital original, The Journal on Technology and Persons with Disabilities 6 (2018) 33–42. [20] L. Moreno, H. Petrie, P. Martínez, R. Alarcon, Designing user interfaces for content simplification aimed at people with cognitive impairments, Universal Access in the Information Society 23 (2024) 99–117. doi:10.1007/s10209- 023- 00986- z. [21] S. Trewin, S. Basson, M. Muller, S. Branham, J. Treviranus, D. Gruen, D. Hebert, N. Lyckowski, E. Manser, Considerations for AI Fairness for People with Disabilities, AI Matters 5 (2019) 40–63.
URL: https://doi.org/10.1145/3362077.3362086. doi:10.1145/3362077.3362086. [22] M. J. Muller, Participatory Design: The Third Space in HCI, in: J. A. Jacko, A. Sears (Eds.), The
Human-Computer Interaction Handbook, CRC Press, 2003, pp. 1051–1068. [23] P. Lane, H. Mckenna, A. Ryan, P. Fleming, Focus group methodology, Nurse Researcher 8 (2001) 45–59. doi:10.7748/nr2001.04.8.3.45.c6157. [24] D. Morgan, M. Spanish, Focus groups: A new tool for qualitative research, Qualitative Sociology 7 (1984) 253–270. doi:10.1007/BF00987314. [25] K. Parahoo, Nursing Research: Principles, Process and Issues., Palgrave Macmillan, United Kingdom, 1997. [26] R. A. Krueger, Focus Groups: A Practical Guide for Applied Research, 2nd ed., Sage Publications,
Thousand Oaks, CA, 1994. [27] K. L. Then, Focus group research, Canadian Journal of Cardiovascular Nursing 7 (1996) 27–31.
URL: https://pubmed.ncbi.nlm.nih.gov/9136312/, pMID: 9136312. [28] J. L. Keelor, N. Creaghead, N. Silbert, T. Horowitz-Kraus, Text-to-Speech Technology: Enhancing Reading Comprehension for Students with Reading Dificulty, Assistive Technology Outcomes and Benefits 14 (2020) 19–35. Available online: https://www.atia.org/atob. [29] K. Wołk, A. Wołk, W. Glinkowski, A cross-lingual mobile medical communication system prototype for foreigners and subjects with speech, hearing, and mental disabilities based on pictograms, Computational and Mathematical Methods in Medicine 2017 (2017) 4306416. doi:10.1155/2017/ 4306416, epub 2017 Nov 2.