DECILE

According to the World Health Organization, approximately 430 million people worldwide suffer from incapacitating hearing loss. In Argentina, nearly a million individuals have permanent hearing limitations, including 64,000 children aged 0 to 14.

Hearing aids and cochlear implants (CI) have significantly improved the hearing abilities of deaf individuals, enabling many children to acquire spoken language skills with appropriate stimulation. However, hearing through a CI differs from typical auditory development, requiring greater effort to comprehend linguistic sounds. 

In Argentina, CIs are usually implanted after the age of one, or even later, due to a lack of information among parents and unstable healthcare policies influenced by budget constraints. As a result, Deaf and Hard of Hearing (DHoH) children whose families do not use sign language (representing about 95% of them), are exposed to oral language at a later stage in childhood. This delay negatively impacts their linguistic, educational, emotional, and social development.

From a linguistic point of view, DHoH children who exclusively access oral language often face difficulties in morphology and syntax, showing delays in utterance length development, omissions or incorrect usage of determiners, prepositions, and inflectional morphology when compared to typically developing children. They also struggle with understanding and producing syntactically complex sentences.

The proposed solution takes into account two significant gaps in the existing research on DHoH children who use oral language:

First, although there are numerous studies available on other languages, evidence regarding linguistic attainment of Spanish-speaking deaf children who use CIs and hearing aids is limited, and most of it comes from data obtained in high-income countries. 

Second, within the DHoH children’s group, a great heterogeneity in performance is usually reported, along with differences in children’s health history, socio-economic and linguistic background. Yet, the connection between these sources of heterogeneity and variations in performance remains poorly understood, and educational intervention programs are typically developed based on protocols that assume homogeneity within the group. Consequently, healthcare and education professionals rely on published results from these programs to make recommendations, leading to an unfortunate chain of misinformed decisions for these children.

DECILE is part of a software suite targeted at children with language difficulties. The app measures children’s performance on several linguistic metrics (receptive vocabulary, accuracy and response time in sentence production and comprehension, accuracy in sentence repetition and intelligibility) together with information about external factors that may affect language development.  

Targeted at the educative community, our team has focused on establishing the key characteristics of our educational software, with a particular emphasis on:

• User-friendly and intuitive interface

• Engaging and interactive content

• Flexibility and scalability

• Accessibility and inclusivity

• Data privacy and security

Working together with schools that target neurodiverse children, the app has adapted school-based experimental tasks to a software experience that is engaging for children and  fast and easy to use for language therapists, teachers and parents. Evaluation tasks are game-based activities that take between 15 to 20 minutes, with a main character that guides children along the different tasks. Educators and practitioners can use this app to gain insights on particular difficulties the child might have, and compare performance across sessions and guide their interventions accordingly.

Teachers and practitioners using the app first create a profile for the child, entering demographic and etiological data if available. They then select the task and can adjust parameters such as response time and the number of response opportunities provided. The child begins the task with the assistance of Popi, a child-guide who provides examples and feedback to ensure understanding before the screening begins.

DECILE is grounded in scientific principles yet offers flexibility. The tasks are designed based on psycholinguistic metrics of Spanish language development but can be adapted to users' needs by modifying input sound and image files. This feature makes it suitable for assessment in other Spanish-speaking countries interested in customizing aspects such as Popi's pronunciation. A demo of one of the comprehension tasks can be found here.

Following each session, all data is securely recorded and stored in a central database, including accuracy rates, response times, and the child's responses in production tasks. During the pilot phase, collected data will be utilized to train machine-learning models to identify language learning profiles among children, enabling practitioners and teachers to implement tailored interventions effectively.

Two children can share the same birth time, location, and socioeconomic background, yet their educational paths may diverge significantly. This disparity becomes even more pronounced when language learning becomes a challenge. For DHoH children, numerous variables can influence language acquisition and subsequent educational achievements. These include:

• Children’s health history and etiology of deafness

From a medical standpoint, the period from birth to CI fitting varies among children, impacting their exposure to oral language at different ages and potentially affecting language acquisition opportunities. Differences in performance may also arise based on the type of cochlear implant used or whether CI fitting is needed for one or both ears. Etiology could also introduce language learning variability. For instance, while deafness runs genetically in some families, it could be the consequence of an episodic disease, or the result of premature birth, leading parents to take action at different stages during language development.

• Children’s linguistic history

The educational paths of DHoH children exhibit significant diversity, largely influenced by parental decisions following their child's diagnosis of deafness. It is essential to recognize that nearly 95% of deaf children are born to hearing families. Consequently, congenitally deaf children raised by sign language-using parents typically exhibit a native proficiency in sign language and often learn Spanish as a secondary language. However, families are often confronted with the decision of whether to learn sign language alongside their child while awaiting CI fitting, or to pursue exclusive oral linguistic education. Even in cases of the latter, CI may not always yield the anticipated results, leading some children to begin learning sign language later on. With this array of options, many DHoH children encounter educational barriers, as they may not be ready to start school with their peers of the same age and are frequently placed in specialized schools.

• Demographic differences 

Demographic variations also influence the linguistic education and subsequent performance of DHoH children, who may belong to families with varying socio-economic backgrounds or live in countries with differing access to universal hearing screening. These factors frequently correlate with parents' access to information, which, in turn, leads to (mis)informed decisions and exacerbates inequalities among deaf children. Consequently, when assessing differences in language achievement to guide policy development, the data collected from DHoH children are usually biased due to this disparity.

Considering these factors, we have developed DECILE, an intuitive software designed to assess a range of linguistic abilities. This state-of-the-art tool aims to meet the demand for comprehensive evaluation, while also adapting to the complexities and individual differences that influence language development. 

With the data gathered through the app, we hope to use machine learning models to aid researchers to understand how various factors may shape language learning paths for these children. Practitioners and educators can then use this insight to understand the individual variability of each child and refine interventions accordingly, fostering more effective support. Ultimately, our long-term goal is that early language assessment and intervention will empower children facing language difficulties to forge brighter futures.

DECILE began to take shape after numerous meetings with school principals who serve DHoH children—an area of keen interest for our team’s research. They were concerned that existing language assessment tools were not specific enough and took too long. They also worried that these tests seemed too clinical, which could make them less appealing to children and affect the results. Additionally, they sought to understand if their intervention and education programs effectively addressed the issues they observed in their students but struggled to measure accurately. Since 2017, our team has been dedicated to collaborating with these schools, aiming to provide practical, research-backed learning solutions. Specifically, three schools focusing on oral education for DHoH students requested the development of a diagnostic tool for assessing their Spanish language abilities. Since then, we have conducted various research studies within schools, using methods from cognitive neuroscience to address questions such as:

• What communication strategies do DHoH children equipped with cochlear implants employ during pedagogical events and how do they differentiate from those used by children who typically develop hearing? What comes first, the ostensive channel or the content of knowledge? 

• What has been described in the literature as typical comprehension patterns in orally educated children with prelingual hearing loss? 

• How do Spanish-speaking deaf children perform in sentence comprehension? And how do they perform in oral production?

• When assessing receptive vocabulary in DHoH children with cochlear implants who speak Spanish, how much weight should be given to their hearing age? 

• How effective are current Spoken Language Processing (e.g., Automatic Speech Recognition) tools when used by DHoH children, and how can they be improved to enhance accessibility? 

• What is the impact of bimodal bilingualism in adults’ performance in tasks related to Executive Functions. To further explore this area, we have engaged three bilingual schools (teaching Argentine Sign Language and Spanish) to participate in additional experimental studies taking place in 2024. Our objective is to understand the consequences of bilingualism on linguistic attainment in these children, considering the significance of Executive Functions for learning.

In 2022, our team received a grant from the Inter-American Development Bank to launch a project using Behavioral Science methods to enhance Policy outcomes. This project focuses on exploring how parents of Deaf and Hard of Hearing (DHoH) children in Argentina make decisions about their children's linguistic education. Our goal is to promote bimodal bilingual (Argentine Sign Language - Spanish) education for these children, encouraging sign language learning among parents, caregivers, and close relatives. The project is a collaboration with the Ministries of Education in Buenos Aires and Mendoza, and is currently being implemented in all public schools serving DHoH children in these cities.

The ongoing research we've conducted has led to the creation of DECILE. Originally developed for Deaf and Hard of Hearing (DHoH) children, our aim is for this tool to eventually benefit other communities dealing with linguistic challenges as well.

Creating DECILE has been an extensive process: before releasing the app, we conducted analog testing of DECILE's materials with Deaf and Hard of Hearing (DHoH) children as well as typically developing children. Our goal was to ensure that our tasks accurately measured their intended targets, following the standards of rigorous psycholinguistic studies. Throughout this process, we refined and enhanced images and instructions to ensure they were accessible and inclusive for all children dealing with linguistic challenges.

In the next phase, we transformed the tasks into interactive app-based games. This involved solving a wide range of problems, including the development and testing of the program's algorithm, designing the aesthetics and defining the characteristics of Popi -the guide within our game—who was selected as the favorite by a group of 30 children of various ages. We also recorded sound files tailored for Cochlear Implant (CI) users. This effort resulted in the creation of two app-based sentence comprehension tasks.

In 2023, DECILE was initially introduced to three primary schools in Buenos Aires, Argentina. We evaluated the comprehension skills of around 200 Deaf and Hard of Hearing (DHoH) as well as typically developing children aged 3 to 11. The results from these tasks align with those from traditional screening methods in terms of accuracy, while also providing additional metrics such as response times and the number of prompts before the child responds.

Throughout this year, we also conducted tests for the materials of a sentence production task and a narrative production task. Additionally, we explored the effectiveness of cutting-edge Automatic Speech Recognition programs in measuring intelligibility among children with Cochlear Implants and hearing aids. Presently, we are integrating these findings to adapt our paper-based production tests, aiming to introduce two oral production tasks to DECILE.

The goal of these assessments is to establish a baseline for quantifying the variability in language learning among Spanish-speaking DHoH children from Argentina, focusing on sentence comprehension and production. In the upcoming phases, our objective is to enhance our interface to offer instant metrics that compare a child's performance across sessions and with peers in the same age group.

Furthermore, with the data collected so far we are working on the implementation of machine learning techniques to develop models capable of predicting the performance of Deaf and Hard of Hearing (DHoH) children in various linguistic tasks. We will evaluate the quality of these models using standard regression metrics. Once trained and evaluated, these models will allow us to examine the influence of different sociological and health-related factors on linguistic performance.  This information can offer valuable insights, aiding in the design of targeted intervention programs and equipping health practitioners, educators, and policymakers with additional guidance to offer timely recommendations to parents of deaf children.

In addition to gaining visibility, joining SOLVE will greatly benefit us in several key ways. First, we're seeking expert guidance in marketing to enhance the appeal of our platform for potential users and investors, boosting our solution to a higher level of growth. This entails reaching out to schools and practitioners across the country who have yet to benefit from our software. On this end, we hope to gain expert advice on how to design DECILE’s business model, and how to make our solution sustainable in time.

Our team aims to expand DECILE's testing to diverse populations facing linguistic challenges, and children from both urban and rural areas, broadening our user base and promoting inclusivity. The latter ones, often underserved in terms of health and education opportunities, stand to gain the most from our platform. Thus, we require training in scalability and assistance in crafting a marketing strategy that caters both practitioners and educators from both urban and rural regions.

Finally, we hope for expert advice on legal matters, particularly in ensuring robust data privacy and security measures for our proposed solution.

DECILE changes language assessment for children with a multifaceted approach. Unlike traditional methods, it offers a comprehensive evaluation through user-friendly, fun and engaging tasks covering various linguistic metrics. 

Its adaptability allows customization to individual needs, fostering inclusivity. By incorporating Artificial Intelligence, DECILE not only provides valuable insights for researchers but also enables targeted interventions. 

This innovative solution has the potential to create positive impacts by setting a new standard for language assessment, inspiring others in the field to adopt similar comprehensive approaches that emphasize in individual outcomes rather than in group homogeneity. 

Ultimately, DECILE could transform the market by elevating the quality of language assessment tools, leading to improved educational outcomes and better support for children facing linguistic challenges.

Educators often highlight the diversity among DHoH children and the challenges that following homogeneous education programs impose. DECILE helps us understand these differences, offering detailed insights about individual information and performance to therapists and teachers. This understanding fosters better school outcomes, boosting confidence in these children and creating opportunities for improved life chances, promoting equity and inclusion.

A Diagram of our solution’s Theory of Change can be found here.

Our solution uses machine learning techniques to build models capable of estimating the performance of DHoH children in different linguistic tasks, including oral vocabulary comprehension and morphosyntactic comprehension and production. The quality of these models will be assessed via standard evaluation metrics for regression, such as the root mean squared error (RMSE). Once these models have been trained and evaluated, they will allow us to analyze the relative impact on the linguistic performance of the different sociological and health-related variables, both in isolation and combined, which may offer cues to better-targeted intervention programmes.

Full time: 4

Part time: 3

Contractors: 2

7

Based at the School of Government at UTDT, the Center for Evaluation of Policies and Evidence-Based Policy-Making (CEPE Di Tella) specializes in applied research aiming to enhance public policies. Since its establishment in 2017, CEPE Di Tella and its researchers have undertaken numerous projects funded by international organizations like the World Bank and UNICEF. Additionally, they've secured grants from philanthropic foundations like the BBVA Foundation and collaborated with institutions like the Center for International Development at Harvard University.

CEPE Di Tella’s Education and Behavioral Sciences Program focuses on the analysis of educational policies using social, political, pedagogical, economic, contextual, behavioral, and neuroscientific determinants of learning using applied research quantitative and qualitative methods. Our program focuses on researching learning and teaching capabilities across the lifespan, with a particular emphasis on populations that may lack equal resources and abilities. Our primary objective is to foster equity and provide equal opportunities for all children.

Since 2017, CEPE Di Tella has been collaborating with schools in Argentina to develop the Learning to Learn Program (LeLePro), targeting DHoH children aged 3 to 11. LeLePro involves evaluating cognitive abilities using a software suite. DECILE assesses oral language comprehension and production, while MetaTom evaluates metacognition and Theory of Mind (ToM). The software also gathers demographic and environmental variables that may impact linguistic development.

In late 2023, MetaTom received a biannual research grant of USD 2000 from Argentina's National Research and Technology Council to develop the app-based evaluation tool. DECILE's development was made possible through funding from the same institution and research funding linked to the DECILE Team Leader's position at Universidad Torcuato Di Tella.

After expanding product accessibility, CEPE Di Tella plans to distribute the app to schools, language therapist practices, and individuals, providing training and talks for compensation to maintain sustainability. They also aim to increase revenue by partnering with other potential investors, like Cochlear Implant companies.