Impacto de un Sistema de Tutoría Inteligente con Reconocimiento Emocional en el Académico Preescolar
Resumen
Los sistemas de tutoría inteligente (STI) representan una herramienta prometedora para la educación temprana; sin embargo, la mayoría ignora la dimensión afectiva del aprendizaje, limitando su capacidad de personalización pedagógica. Esta omisión es especialmente crítica en educación preescolar, donde los niños de 3 a 5 años dependen de la regulación emocional externa para sostener el compromiso cognitivo. Desde el marco de la computación afectiva y la teoría de la carga cognitiva, se hipotetiza que un STI sensible a las emociones puede reducir la carga extrínseca asociada a la frustración y potenciar el aprendizaje efectivo. Este estudio evaluó el impacto de un STI con módulo de reconocimiento emocional en el rendimiento académico y el compromiso de niños preescolares. Mediante un diseño cuasi-experimental con 160 participantes (GE=80, GC=80), se compararon tres métricas principales: tasa de respuestas correctas, número de intentos y tiempo de respuesta, en una batería de tareas matemáticas estructurada en cinco niveles de dificultad progresiva, administrada durante cinco semanas con tres sesiones semanales. El GE interactuó con un STI equipado con reconocimiento facial automático (OpenCV + red neuronal convolucional, precisión ~88%) que adaptaba las actividades en función de la emoción detectada (felicidad, neutralidad, enojo). El GC empleó el mismo sistema con el módulo afectivo desactivado. Los resultados indicaron que el GE superó significativamente al GC en todas las métricas evaluadas: mayor tasa de respuestas correctas (78.5% vs. 72.3%, p=.005, d=0.45), menor número de intentos (1.42 vs. 1.68, p=.001), tiempo de respuesta más rápido (18.5 s vs. 22.8 s, p<.001, d=0.58) y mayor engagement observacional (+12%, p<.001, d=0.71). Adicionalmente, la ventaja del GE se amplificó en los niveles de mayor dificultad cognitiva, sugiariendo que el andamiaje afectivo es más valioso ante tareas que generan mayor carga cognitiva. Se concluye que la incorporación de reconocimiento emocional en STI mejora significativa y prácticamente el rendimiento académico y el compromiso en educación preescolar, validando la integración de la dimensión afectiva como componente esencial en el diseño de tecnologías educativas para la primera infancia.
Descargas
Citas
Agarwal, M., & Mostow, J. (2020). Semi-supervised affect detection from speech in real classroom environments. Semi-supervised affect detection from speech in real classroom environments., 712-724. Transactions on Affective Computing, 11(4). https://doi.org/https://doi.org/10.1109/TAFFC.2018.2854299
Eisenberg, N., Spinrad, T. L., & Eggum, N. D. (2010). Emotion-related self-regulation and its relation to children’s maladjustment. Emotion-related self-regulation and its relation to children’s maladjustment., 495-525. Annual Review of Clinical Psychology, 6. https://doi.org/https://doi.org/10.1146/annurev.clinpsy.121208.131208
Ekman, P. (1992). An argument for basic emotions. An argument for basic emotions., 169–200. Cognition & Emotion, 6(3–4). https://doi.org/https://doi.org/10.1080/02699939208411068
Gordon, G., Spaulding, S., Westlund, J. K., Lee, J. J., Plummer, L., Martinez, M., Das, M., & Breazeal, C. (2016). Affective personalization of a social robot tutor for children’s second language skills. Affective personalization of a social robot tutor for children’s second language skills., 3951-3957. In Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence.
Li, J. (2025). Multimodal emotion recognition in online learning for children: A meta-analysis of effectiveness and implementation factors. Multimodal emotion recognition in online learning for children: A meta-analysis of effectiveness and implementation factors. Computers & Education, 210, 104952. https://doi.org/https://doi.org/10.1016/j.compedu.2024.104952
Li,, S., & Deng, W. (2020). Deep facial expression recognition: A survey. Deep facial expression recognition: A survey, 1195–1215. IEEE Transactions on Affective Computing, 13(3). https://doi.org/https://doi.org/10.1109/TAFFC.2020.2981446
Lin, H.-C. K., Su, S. -H., Chao, C. -J., Hsieh, C. -Y., & Tsai, S. -C. (2022). Integration of eye-tracking and affective computing technologies for an intelligent tutoring system: Usability evaluation. Integration of eye-tracking and affective computing technologies for an intelligent tutoring system: Usability evaluation., 474-499. Journal of Educational Computing Research, 60(2). https://doi.org/https://doi.org/10.1177/07356331211037145
Liu, Y. (2025). Multimodal adaptive virtual learning environment for early childhood education: Design and empirical validation. Multimodal adaptive virtual learning environment for early childhood education: Design and empirical validation., 245-268. Educational Technology Research and Development, 73(1). https://doi.org/https://doi.org/10.1007/s11423-024-10387-2
Mollahosseini, A., Hasani, B., & Mahoor, M. H. (2019). AffectNet: A database for facial expression, valence, and arousal computing in the wild. AffectNet: A database for facial expression, valence, and arousal computing in the wild., 18–31. IEEE Transactions on Affective Computing, 10(1). https://doi.org/https://doi.org/10.1109/TAFFC.2017.2740923
Park, H. W., Rosenberg-Kima, R., Rosenberg, M., Gordon, G., & Breazeal, C. (2019). Growing growth mindset with a social robot peer. Growing growth mindset with a social robot peer., 137-145. In Proceedings of the 2019 ACM/IEEE International Conference on Human-Robot Interaction. https://doi.org/https://doi.org/10.1109/HRI.2019.8673077
Parkhi, O. M., Vedaldi, A., & Zisserman, A. (2015). Deep face recognition. Deep face recognition, 41.1–41.12. Proceedings of the British Machine Vision Conference (BMVC). https://doi.org/https://doi.org/10.5244/C.29.41
Pekrun, R., Goetz, T., Titz, W., & Perry, R. P. (2002). Academic emotions in students’ self-regulated learning and achievement: A program of qualitative and quantitative research. Academic emotions in students’ self-regulated learning and achievement: A program of qualitative and quantitative research., 91-105. Educational Psychologist, 37(2). https://doi.org/https://doi.org/10.1207/S15326985EP3702_4
Picard, R. W. (1997). Affective computing. Affective computing. MIT Press.
Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive load during problem solving: Effects on learning., 257-285. Cognitive Science, 12(2). https://doi.org/https://doi.org/10.1207/s15516709cog1202_4
Viola, P., & Jones, M. J. (2004). Robust real-time face detection. Robust real-time face detection, 137–154. International Journal of Computer Vision, 57(2),. https://doi.org/https://doi.org/10.1023/B:VISI.0000013087.49260.fb
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Eds.). Mind in society: The development of higher psychological processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Eds.). Harvard University Press.
Wagner, S. (2025). AI-powered robots for emotional development in preschool education: A systematic review of design principles and learning outcomes. AI-powered robots for emotional development in preschool education: A systematic review of design principles and learning outcomes., 112-128. Early Childhood Research Quarterly, 70,. https://doi.org/https://doi.org/10.1016/j.ecresq.2024.09.003
Woolf, B. P., Burleson, W., Arroyo, I., Dragon, T., Cooper, D., & Picard, R. (2010). Affect-aware tutors: Recognising and responding to student affect. Affect-aware tutors: Recognising and responding to student affect, 129-164. International Journal of Learning Technology, 4(3-4). https://doi.org/https://doi.org/10.1504/IJLT.2009.028804
Derechos de autor 2026 Juan Pablo Diaz Fierro, Francisco Javier Gutiérrez Mata
Esta obra está bajo licencia internacional Creative Commons Reconocimiento 4.0.
.png)
.png)
1.png)
