The Limits of Traditional Grading

Progress in maths in schools has been communicated through grades, percentages, and marks (Link et al., 2025; Normann et al., 2023). These numbers often become the dominant story of a student’s ability, even though they offer only a narrow snapshot of a much richer learning journey. When we consider students who are at risk of dyscalculia, meaning those who show persistent difficulties with early number concepts, counting and foundational maths skills, the limitations of grading become even clearer (Hornos-Arias et al., 2025; Lamb et al., 2024). A single number or letter cannot capture the many abilities that contribute to mathematical development, and it rarely reflects the nuanced ways children think, process and build understanding.

Understanding Early Indicators of Dyscalculia

Children who are at risk of dyscalculia often demonstrate subtle but important patterns (JellyJames Publishing Ltd., b; Lamb et al., 2024; Hornos-Arias et al., 2025). They may rely on concrete counting long after peers have moved on, feel uncertain when recognising number patterns, experience difficulty with quantities, mix up sequences or struggle to retain number facts. These early indicators do not reflect low potential or achievement. Instead, they show that the learner may need more structured support, more time to internalise number relationships and more multisensory experiences to secure the building blocks of mathematical thinking (Esmail, 2020).

How Dynamo Maths Responds to Early Learning Needs

Dynamo Maths recognises the importance of catching these early signals (JellyJames Publishing Ltd., a, b). Instead of waiting for a formal diagnosis, which may arrive too late to prevent wider difficulties, Dynamo Maths focuses on identifying developmental gaps as soon as they appear. Its structured dyscalculia assessment does not simply assign a single grade (JellyJames Publishing Ltd., c; Esmail, 2020). It reveals the conceptual foundations a student has already formed, those that are emerging, and those that remain uncertain. This developmental lens is essential for children at risk of dyscalculia because it shifts the conversation from performance to understanding and insight.

A Changing Educational Landscape

The educational landscape is undergoing a significant transformation. Schools are increasingly embracing personalised learning, early intervention and the idea that meaningful progress cannot be reduced to a grade (Holmes et al., 2019; Panqueban & Huincahue, 2024; UNESCO, 2023). At the same time, artificial intelligence is becoming a deeper part of the learning environment (Holmes et al., 2019; UNESCO, 2023). This combination of AI and developmental assessment could help us move from a grading culture to a more compassionate model of growth.

Why Grades Can’t Capture the Full Learning Journey

Grading was created for a system in which all students were expected to progress through mathematics at the same pace (Link et al., 2025; Normann et al., 2023). However, children who are at risk of dyscalculia remind us that learning does not unfold uniformly. A single test score cannot show if a child has internalised the concept of quantity, can subitise small groups of objects, their working memory is overloaded or if symbolic representations feel too abstract (Lamb et al., 2024; Hornos-Arias et al., 2025). Grades cannot show the frustration of a child who tries but cannot decode the patterns others see with ease. They cannot explain why a learner appears confident one day and uncertain the next. The learning profile is far more detailed, and this is exactly where AI is beginning to make a significant impact (Holmes et al., 2019; Panqueban & Huincahue, 2024). Artificial intelligence is not designed to replace teachers or specialist tools like Dynamo Maths. It supports and enhances them (Holmes et al., 2019; Kim, 2024; UNESCO, 2023).

AI Inside the Dynamo Maths Environment

Within the Dynamo Maths environment, AI strengthens the precision of early identification (JellyJames Publishing Ltd., d; Panqueban & Huincahue, 2024). It helps teachers recognise emerging risks more quickly and suggests pathways matched to each learner’s developmental stage. A child who struggles to recognise number patterns can be guided back to visual experiences. A learner who finds it difficult to retain number facts can be supported through repeated multisensory activities until the concept becomes secure (Hornos-Arias et al., 2025). The learning process becomes dynamic and personalised. It is shaped by the learner rather than imposed upon them.

Early Intervention as a Natural Part of Learning

Early intervention becomes natural and integrated rather than reactive (Hornos-Arias et al., 2025; Lamb et al., 2024). Instead of waiting until a grade reveals a deeper issue, AI can highlight areas of difficulty as they arise. Teachers can introduce targeted support through Dynamo Maths activities to ensure misconceptions do not become ingrained (JellyJames Publishing Ltd., d; Esmail, 2020). This timely support helps learners build confidence, resilience and a positive sense of mathematical identity long before difficulties turn into long-term disengagement.

The Evolving Role of the Teacher

As AI supports identification and adaptation, the teacher’s role becomes even more significant (Holmes et al., 2019; Kim, 2024; UNESCO, 2023). Teachers become mentors who understand that mathematical growth is both emotional and cognitive. For learners at risk of dyscalculia, a supportive relationship with a teacher can completely transform their self-belief (Roorda et al., 2011). Teachers interpret AI insights with human understanding, recognising that data reveals patterns but that each learner brings feelings, experiences and personal context that shape their journey (Holmes et al., 2019; Roorda et al., 2011). Teachers provide the human connection that brings learning to life. AI can guide a pathway, but only a teacher can nurture curiosity, reduce anxiety and celebrate every step forward. Their role becomes one of guide, encourager and advocate. They help learners build secure foundations and develop a positive relationship with mathematics.

A More Human, More Personalised Future for Maths Learning

The future of maths learning is not defined by a single grade. It is defined by understanding. Developmental profiles replace one-dimensional scores. Personalised pathways replace one-size-fits-all instruction. With AI enabling early identification and Dynamo Maths providing structured developmental support, teachers are free to illuminate each learner’s journey with expertise, creativity and compassion (Hornos-Arias et al., 2025; Panqueban & Huincahue, 2024; JellyJames Publishing Ltd., a).

References

Esmail, K. (2020). Standardisation and validation document [Technical report]. JellyJames Publishing Ltd.

Holmes, W., Bialik, M., & Fadel, C. (2019). Artificial intelligence in education: Promises and implications for teaching and learning. Center for Curriculum Redesign.

Hornos-Arias, J., Grau, S., & Serra-Grabulosa, J. M. (2025). Early detection and intervention of developmental dyscalculia using serious game-based digital tools: A systematic review. Information, 16(9), 787.

JellyJames Publishing Ltd. (a). Dynamo Maths – Overview.

JellyJames Publishing Ltd. (b). Dyscalculia.

JellyJames Publishing Ltd. (c). Validation & standardisation.

JellyJames Publishing Ltd. (d). Personalised learning & intervention for dyscalculia.

Kim, J. (2024). Leading teachers’ perspective on teacher–AI collaboration in education. Education and Information Technologies, 29, 8693–8724.

Lamb, S., Krieger, F., & Kuhn, J.-T. (2024). Delayed development of basic numerical skills in children with developmental dyscalculia. Frontiers in Psychology, 14, 1187785.

Link, L. J., Leeson, M. P., & Anthes, J. R. (2025). The problem with traditional grading systems. Educational Leadership, 82(8).

Normann, D.-A., Sandvik, L. V., & Fjørtoft, H. (2023). Reduced grading in assessment: A scoping review. Teaching and Teacher Education, 135, 104336.

Panqueban, D., & Huincahue, J. (2024). Artificial intelligence in mathematics education: A systematic review. Uniciencia, 38(1).

Roorda, D. L., Koomen, H. M. Y., Spilt, J. L., & Oort, F. J. (2011). The influence of affective teacher–student relationships on students’ school engagement and achievement. Review of Educational Research, 81(4), 493–529.*

UNESCO. (2023). Generative AI and the future of education. UNESCO.

Karima Esmail

Co-founder of Dynamo Maths, an award-winning, research and evidence-based inclusive programme proven to support students with dyscalculia and those not meeting age-related expectations.

I have 15 years of experience as a senior lecturer at the University of Hertfordshire. More recently, I undertook research at University College London, which inspired the creation of Puffin Maths for the deaf and hard of hearing, which now provides access to the National Curriculum using British Sign Language (BSL). Puffin Maths was ‘Highly Commended’ at BETT 2022.

I have also co-authored Dynamo Post 16 Number Sense Profiler and Dyscalculia Self Perception Indicators Questionnaire.