Children’s cognitive assessment in digital age

Dear Sir,

Pediatric neuropsychological assessment is at a turning point in North Africa. The rapid penetration of mobile internet, reaching almost half the regional population by 2023 – an exponential growth compared to the previous decade^[1] – opens unprecedented prospects for the digitization of diagnostic tools. The appeal is undeniable: Digital platforms promise standardized administration, near-instant automated scoring, potential long-term cost reduction, and above all, the capture of fine-grained behavioral data (latency times, oculomotor trajectories, and micro-movements) that largely elude traditional paper-and-pencil tests.^[2] Tele-assessment, brought to the fore by pandemic necessity, adds the promise of wider access to services, particularly in regions under-serviced by specialists.^[3]

Faced with this transformative potential, the temptation is great to quickly adopt the digital solutions available on the international market. However, a technological rush, ignoring deep-rooted contextual specificities, risks not only missing its target but also exacerbating existing inequalities. The crucial question is not so much if we should digitize child cognitive assessment in North Africa, but how to do it in a responsible, ethical, and culturally relevant way.

Emerging regional initiatives are charting a promising course. In Tunisia, work has validated the use of a digital battery to effectively distinguish cognitive profiles in adults, including in the early stages of Alzheimer’s disease, demonstrating the clinical relevance of these tools in the local context.^[4] In Egypt, the rigorous validation of the harmonized cognitive assessment protocol on a large sample paves the way for country-specific longitudinal studies of cognitive aging.^[5] Although these examples focus on adults, they illustrate an essential approach: Giving priority to the validation and production of local standards, rather than the uncritical import of instruments developed elsewhere. Unfortunately, such initiatives are still all too rare in the Maghreb, especially for the pediatric population.

The urgency of a contextualized approach is underlined by the proven risks of cultural bias. A study carried out in Morocco on the use of Raven’s colored progressive matrices is particularly eloquent: The direct application of British norms led to up to 15% of typically developing Moroccan children being wrongly identified as potentially deficient.^[6] This example crudely illustrates the non-verbality of a test in no way guarantees its universality. Problem-solving strategies, familiarity with certain types of abstract stimuli, or even the interpretation of instructions can be profoundly influenced by cultural and educational background. Digitization, with its specific graphic interfaces, interactive metaphors, and possible playful components, can introduce new layers of bias if it is not designed with extreme cultural sensitivity.^[2] Digital literacy itself, unevenly distributed, becomes a potential confounding variable.^[7]

So how can we move forward in a rigorous and fair manner? The guidelines for technology-based assessment, published jointly by the International Test Commission and the Association of Test Publishers in 2022, offer a valuable methodological framework.^[8] They emphasize key principles such as universal design (aiming for the widest possible accessibility), the need for differential item functioning analyses to detect and correct potential biases between cultural or linguistic groups, the importance of robust protocols for remote administration, and the implementation of offline solutions to overcome inequalities in connectivity. These guidelines should form the minimum foundation for any initiative to develop or deploy digital testing in the region.

Cultural adaptation must therefore be at the heart of the process. It involves much more than simple translation: Conceptual revision of items, adaptation of visual and auditory stimuli, rigorous linguistic validation considering the different languages and dialects spoken (standard Arabic, Darija, Amazigh languages), and above all, calibration on large local samples representative of socio-economic and geographical diversity. This colossal task calls for close interdisciplinary collaboration between psychometricians, clinical neuropsychologists, linguists, anthropologists, computer scientists, and local educators.

The ethical dimension is just as crucial. The collection of sensitive cognitive data, especially from minors, requires impeccable governance. The World Health Organization’s global strategy on digital health (2020–2025) provides a general framework,^[9] but its practical application requires systematic endorsement by local ethics committees, appropriate informed consent procedures (involving parents and, depending on age, the child’s assent), robust cybersecurity and anonymization guarantees, and respect for the right to digital oblivion. Artificial intelligence, which offers fascinating prospects for adaptive testing and the identification of digital biomarkers, must itself be developed and audited with vigilance to avoid encoding or amplifying existing societal biases.^[7] The transparency of algorithms and the representativeness of training data are non-negotiable prerequisites.

Given the scale of the task, a fragmented approach would be ineffective. Setting up a Maghreb research consortium dedicated to children’s digital cognitive assessment could enable us to pool resources, share best practices, and develop harmonized but locally validated tools and standards. Multicenter pilot studies comparing paper and digital formats in different contexts (urban/rural, public/private) are essential. The development of open-source platforms, compliant with international standards and adapted to local constraints (low bandwidth, offline operation), would promote wider and more equitable dissemination.

In conclusion, the digitization of cognitive assessment in North Africa is a necessity to meet the needs of a young and increasingly connected population. However, this transition must not be at the expense of scientific rigor and equity. It offers us a unique opportunity to move beyond the acritical import model and co-construct truly adapted tools that respect the cultural richness of the region and best serve the development of each child. This is a major scientific, clinical, ethical, and political challenge, which our community must seize with responsibility and ambition.