SAN 2026

V-55

Cognition, Behavior, and Memory

Dynamic Difficulty Adjustment in Child Cognitive Training: Converging Evidence Across Two Countries

Gabriel Osvaldo Paz1,2, Martina Boscolo1,2, Luis Bustamante3, Daniela Macario-Cabral3, Diego Slezak2,3, Andrea Paula Goldin1,2

1. Universidad Torcuato Di Tella. Escuela de Negocios. Centro de Inteligencia Artificial y Neurociencia (CIAN). Laboratorio de Neurociencia. CONICET.
2. Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Buenos Aires 1428, Argentina.
3. Universidad de Buenos Aires, Facultad de Ciencias Exactas y Naturales, Departamento de Computación, Laboratorio de Inteligencia Artificial Aplicada (LIAA).


Presenting Author:

Gabriel

Paz

gabrielpaz.uba@gmail.com

For the past 15 years, we have used the web-based cognitive assessment and training platform Mate Marote with children aged 4 to 8. An initial school-based study in Navarra, Spain, showed that a dynamic difficulty progression model effectively enhanced attention. A second study, conducted again in the same school, evaluated a revised version of this progression designed to overcome its limitations. In the present work, we compare both models. Both approaches adapt to the player’s initial level; however, the original model slows progression after reaching a balance point, whereas the revised model continuously adjusts task difficulty. We hypothesized improvements in executive functions from pretest to posttest, with greater gains under the revised model. Results showed attentional improvements in both groups, while only the revised model produced additional gains in logical reasoning, cognitive flexibility, and working memory. These findings were replicated with a local school sample in Tandil, Buenos Aires, Argentina.