Big ideas.
Brighter futures.

Nova Learning Notes · Research commentary · Updated 23 September 2026

Technology can make high-quality learning support more available, more responsive and easier to revisit. A learner can return to an explanation, annotate a diagram or receive guidance while an idea is still taking shape. By September 2026, research offers an increasingly informative account of where these benefits arise: purposeful practice, useful feedback, accessible teaching and well-supported educators. The strongest educational case connects these capabilities to the work of learning itself.

Making the learning process visible

A useful way to understand educational technology is through the opportunities it creates for interaction. Written working gives a teacher something to discuss. A saved question identifies a point of uncertainty. A sequence of attempts can show how an explanation has been understood. Our interpretation is that digital tools are especially valuable when they help learners articulate their thinking and help teachers choose an appropriate next step.

Guidance that responds to the learner

Kestin and colleagues’ university trial suggests that carefully designed tutoring can support understanding through questions, feedback and a manageable sequence of steps. Its short duration and specific subject setting matter when interpreting the findings.¹ For Nova, this informs a design aim: connect independent practice with a teacher’s knowledge of each learner. The research is external to Nova; our own learning impact remains to be evaluated.

Handwriting within a digital classroom

In an EEG study of 36 university students, van der Weel and van der Meer found more extensive brain connectivity during digital-pen handwriting than typing. Participants used a Microsoft Surface Studio, so this was not an iPad trial or a test of attainment.² A separate 2026 study comparing ink pens, digital pens and keyboards found no significant difference in immediate test performance.³

Our interpretation is to preserve a choice of writing, drawing and typing. An iPad with Apple Pencil can support handwritten notes and annotation in compatible apps.⁴ This offers a practical way to keep handwriting in digital study, while allowing the task, the learner and the teaching approach to guide the choice of device.

Examples of adoption

Maryville University’s Digital World programme provides an iPad to all traditional, full-time undergraduates, illustrating how devices can be integrated into everyday study.⁵ Jisc’s survey shows positive student perceptions, alongside continuing needs for reliable connectivity and suitable devices.⁶ Access and teaching support deserve the same attention as the technology itself.

What the September 2026 evidence adds

A study published in Nature on 9 September 2026 reports learning gains from phone-based tutoring across five randomised trials during school disruption. It offers a timely example of technology extending access to personal instruction, even through familiar, relatively simple devices.⁸ The practical significance is the connection it makes possible: a learner, a suitable task and someone who can respond.

In England, the DfE and EEF launched an EdTech Testbed in May 2026 to strengthen the evidence available to schools, including for generative AI products. The EEF’s programme also examines adaptive learning, feedback and teacher workload.⁹ This makes careful evaluation part of a positive route to adoption: schools can identify the teaching needs a tool serves and review how well it serves them in practice.

Connecting research to everyday teaching

These findings support a constructive ambition for digital education: more opportunities to practise, more continuity between sessions and more useful information for teachers. For Nova, they inform the design of connected learning records, revision routines and teacher oversight. Our own impact will be assessed with the schools and learners using Nova. The following notes explore feedback, digital handwriting and teaching support in more detail.

References

1. Kestin et al. (2025). AI tutoring and in-class active learning: a randomised controlled trial. Scientific Reports, 15, 17458. ↗
2. Van der Weel & van der Meer (2024). Handwriting and brain connectivity: a high-density EEG study. Frontiers in Psychology, 14, 1219945. ↗
3. Shin et al. (2026). Taking notes with different writing devices influences learning processes but not performance. Frontiers in Education. ↗
4. Apple. Draw and handwrite with Markup on iPad. Device feature documentation. ↗
5. Maryville University. Digital World programme and iPad provision. Institutional account. ↗
6. Jisc (2025). UK higher education students’ digital experience insights, 2024/25. ↗
7. xAI (11 December 2025). El Salvador education partnership announcement. ↗
8. Angrist et al. (9 September 2026). Mobile education builds resilience during shocks in five countries. Nature, 657, 689–696. ↗
9. Education Endowment Foundation. EdTech research programme and the 2026 DfE-EEF Testbed. ↗

Nova Learning Notes · Research commentary · Updated 23 September 2026

Digital learning creates an opportunity to offer support at the moment a learner needs it. The educational benefit comes from what happens next: noticing a misconception, trying a different approach and explaining why it works.

Feedback with a clear next step

The EEF’s evidence synthesis finds positive effects from feedback delivered by teachers, peers and digital systems. Helpful feedback makes the task or strategy clearer and gives the learner information they can use to improve. Its value depends on both its quality and the opportunity to act on it.¹ A digital explanation can make this sequence easier to revisit, allowing a learner to pause and work through one step at a time.

Encouraging the learner to think

Kestin and colleagues’ 2025 Harvard study compared a structured AI tutor with active classroom teaching across two undergraduate physics lessons. The tutoring condition produced higher learning gains. The intervention was deliberately designed around learning principles, including manageable steps and feedback.² The finding supports further study of this approach in other settings; it does not establish the same effect for GCSE pupils.

Connecting independent work to teaching

Repeated questions can offer a productive starting point for a lesson. A teacher might ask a pupil to show where their method changed, or use a shared difficulty to plan a short explanation for the class. Our design aim for Nova is to make that conversation easier through useful learning records and clear next steps.

References

1. Education Endowment Foundation. Feedback evidence synthesis, Teaching and Learning Toolkit. ↗
2. Kestin et al. (2025). Randomised trial of AI tutoring in undergraduate physics. Scientific Reports, 15, 17458. ↗

Nova Learning Notes · Research commentary · Updated 23 September 2026

A digital classroom can retain the familiar acts of writing, sketching and working something out. A stylus gives learners a way to show a calculation, connect ideas in a diagram and annotate a text while keeping their work available to revisit.

What handwriting research tells us

A 2024 EEG study with 36 university students found more extensive connectivity during digital-pen handwriting than typing. It examined neural activity using a Microsoft Surface Studio, rather than measuring exam gains.¹ This offers a reason to take the physical process of writing seriously within a digital learning environment.

A useful 2026 update

Shin and colleagues’ April 2026 study compared ink pens, digital pens and keyboards in note-taking tasks with 33 undergraduates. Brain-activity measures differed by device and strategy, while immediate test scores did not differ significantly.² Our interpretation is to give learners purposeful choices: sketch a process, write a question in their own words or type a structured summary, according to the task.

Bringing the pencil into digital study

Apple’s iPad documentation describes handwriting and drawing in Markup with a finger or a compatible Apple Pencil.³ These capabilities allow a teacher to plan activities that combine a digital resource with visible written reasoning. Device features provide the medium; the activity determines how learners use it.

References

1. Van der Weel & van der Meer (2024). EEG comparison of handwriting and typing. Frontiers in Psychology, 14, 1219945. ↗
2. Shin et al. (17 April 2026). Writing devices, note-taking strategies and immediate learning performance. Frontiers in Education, 11, 1799651. ↗
3. Apple. iPad User Guide: handwriting and drawing with Markup. ↗

Nova Learning Notes · Research commentary · Updated 23 September 2026

Some of technology’s most practical educational benefits are found around the lesson: preparing a useful example, maintaining contact with a learner and making resources easier to access. Research offers encouraging examples of each, with implications for how schools support their staff.

Supporting lesson preparation

An EEF-funded trial, evaluated by NFER and published in December 2024, involved 259 teachers across 68 English secondary schools. Teachers using ChatGPT with an implementation guide spent an average of 56.2 minutes per week on the measured Year 7 and 8 science preparation tasks, compared with 81.5 minutes in the comparison group: a 31% reduction. A blinded review of sampled resources found no apparent reduction in quality.¹

This was a saving within the preparation activities studied, rather than a reduction in teachers’ total working week. It suggests a practical opportunity to support resource creation while teachers retain responsibility for checking accuracy, adapting examples and deciding what their pupils need.

Personal instruction through familiar devices

Angrist and colleagues’ Nature paper, published on 9 September 2026, reports five randomised trials in India, Kenya, Nepal, the Philippines and Uganda. Targeted tutoring by telephone improved learning during school disruption; text-message-only approaches had mixed results.² The study concerns emergency education and foundational learning. Its wider relevance is the potential of accessible technology to sustain a teaching relationship.

Designing for participation

Jisc’s 2024/25 higher education survey recorded positive digital experiences alongside continuing needs for suitable devices and reliable connectivity.³ For schools, our interpretation is to plan access and teaching together. Clear instructions, readable materials and a familiar route to asking for help can make a digital resource easier to use consistently.

References

1. EEF / NFER (2024). Teacher Choices evaluation of ChatGPT-assisted science lesson preparation. ↗
2. Angrist et al. (9 September 2026). Five-country trials of mobile education during disruption. Nature, 657, 689–696. ↗
3. Jisc (2025). UK higher education students’ digital experience survey, 2024/25. ↗