The 10 Most Important Education Research Findings of 2025 (And What They Mean for Physics Students in Singapore)

Every year I set aside time to read through the most significant education research published in the past twelve months. Not the headlines — the actual studies. The ones that tell you something real about how students learn, what holds them back, and what teachers and tutors can do differently.

2025 was a particularly rich year. The research covered everything from cell phone bans in classrooms to the surprising truth about AI and student writing, from brain breaks and attention spans to the neuroscience of handwriting. And running through almost all of it was a theme I have been talking about with physics students in Singapore for years: the gap between knowing something and actually understanding it.

Here are the ten findings that mattered most — and what each one means for how we teach and learn physics.

1. Cell Phone Bans in Classrooms Dramatically Improve Student Grades

The largest study ever conducted on cell phone bans in classrooms is in, and the results are unambiguous.

Researchers installed over 1,000 wooden storage boxes across 10 college campuses in India and ran a randomised controlled trial with nearly 17,000 students. For one full semester, half the students attended class with their phones. The other half deposited them in the boxes before sitting down.

The phone-free classroom group showed substantially better academic outcomes, especially among newer and weaker students. Phone bans, the researchers concluded, can meaningfully improve student grades and narrow achievement gaps. A separate 2025 study of Florida high schools reported similar improvements alongside significant reductions in unexcused absences.

By the end of 2025, 22 US states had passed new laws restricting phone use in schools.

What this means for physics students in Singapore: Distraction is the silent enemy of deep learning. Physics, more than almost any other subject, requires sustained focus — the kind of concentrated attention that smartphone notifications systematically destroy. If you are serious about understanding electricity and magnetism, or working through a quantum mechanics problem set, the phone needs to be physically out of reach. Not silent. Not face-down. Gone.

We explore how focused, distraction-free learning shapes physics performance in our post on why physics students in Singapore know physics terms but cannot use them — a pattern that is directly worsened by fragmented attention during study.

2. The Secret to Solving Maths and Physics Word Problems Is Not Highlighting

If you have a student who reads a physics word problem three times, highlights the numbers, and still gets stuck — this research explains exactly why.

After analysing 1,000 solutions to middle school maths word problems, researchers found that simple highlighting — the most common student technique — was only marginally helpful on its own. What actually worked was using highlighting as a first step in a broader strategy that included sketching diagrams, categorising information, and annotating with arrows and labels.

Students who used these organisational and elaborative note-taking strategies were 29 percent more likely to solve a problem correctly than those who only highlighted. A separate 2025 study found that skilled problem solvers mentally "offload" information onto sketch pads and margin notes — freeing up working memory to focus on the relationships between variables rather than trying to hold everything in their head at once.

What this means for physics students in Singapore: This is exactly how strong physics reasoning works. When I watch a student attempt an O-Level Physics problem on forces or pressure, the ones who draw free body diagrams before touching the numbers almost always outperform the ones who jump straight to equations. The diagram is not decoration — it is thinking made visible.

For a deeper look at how we build this kind of structured problem-solving into our teaching approach, visit our methodology page.

3. 90-Second Brain Breaks Every 10 Minutes Boost Student Attention by Up to 76%

Here is the uncomfortable truth about student attention spans: researchers detected the first signs of declining focus just five minutes into a lecture. After that, attention fell steadily for the rest of the lesson.

A 2025 study enrolled 253 university students in 90-minute lectures structured in two ways: one group received a 90-second microbreak every 10 minutes, while the other group got a single 10-minute break at the halfway point. The microbreak activities were deliberately simple — closing eyes, chatting briefly with a classmate, stretching, or drinking water.

The results were striking. Students in the microbreak group outperformed the control group by up to 76 percent at multiple points throughout the lecture.

The researchers' conclusion is worth taking seriously: instead of trying to overcome the limits of human attention, educators should acknowledge those limits and design lessons around them.

What this means for physics students in Singapore: The standard advice to "study for three hours straight" is neurologically counterproductive. Chunking study sessions into 10-15 minute focused blocks with brief, genuine breaks produces better retention than grinding through fatigue. This is especially important for demanding topics like H2 Physics, where students must hold multiple concepts in working memory simultaneously.

Our A-Level H2 Physics tuition sessions are deliberately structured to account for this — with pacing that allows consolidation, not just coverage.

4. Handwriting Produces Dramatically Better Learning Than Typing

This one surprised even me when I first read it — and I have been tracking literacy and learning research for years.

In a 2025 study, researchers asked 5-year-old pre-readers to learn new letters and words through either handwriting or typing, then tested how well they could name, write, and decode them. The results were not close.

Handwriters achieved 92 percent accuracy in naming letters; typers achieved 75 percent
For writing letters from dictation, handwriters succeeded 64 percent of the time; typers managed only 28 percent

Neuroscience studies of older students show the same pattern. When students write words by hand rather than typing them, brain scans reveal deeper learning activity — traces of more integrated, durable encoding. A separate 2025 study confirmed that handwriting is an important tool for learning and memory retention across all age groups, including secondary and JC students.

What this means for physics students in Singapore: When you type your physics notes, you process words. When you write them by hand, you process meaning. The physical act of forming letters and equations by hand engages more of the brain — and for subjects like physics, where understanding the relationship between variables matters more than memorising the formula, that deeper engagement is enormously valuable.

Write out your derivations. Sketch your diagrams by hand. Copy key equations rather than photographing them. This is not nostalgia — it is neuroscience.

5. Helping Students Too Quickly Destroys Their Confidence and Independent Thinking

This finding has direct implications for how I approach every session with a struggling physics student.

A 2025 research review found that when adults intervene too quickly to help a struggling child, the child interprets it as a signal that the problem is beyond their ability. As early as age five, children who received unsolicited adult help became less motivated to persist on difficult tasks. Among 6 to 11-year-old girls, receiving help they had not asked for left them feeling "less smart."

The researchers' recommendation: instead of rescuing students from confusion, offer hints and guiding questions that point them in the right direction without doing the thinking for them.

What this means for physics students in Singapore: The moment just before understanding — when a student is frustrated and on the verge of giving up — is often the moment when the most durable learning happens. A good physics tutor does not jump in with the answer. They ask: "What do you know for certain? What does Newton's Second Law tell you about the forces here? What would happen if the mass doubled?"

This is the philosophy behind our Multiple Intelligences teaching methodology — building independent thinkers who can navigate unfamiliar problems, not students who need to be rescued every time the question changes.

6. AI Can Handle Special Education Paperwork Without Sacrificing Quality

This finding is less about student learning and more about what happens when teachers are freed from administrative burdens — which directly affects the quality of teaching students receive.

In a 2025 study, experienced special education teachers wrote Individual Education Plan (IEP) goals from scratch, then used ChatGPT to generate IEP goals based on the same student information. Researchers evaluated both sets on six quality dimensions: clarity, measurability, timeliness, and others.

The result: no statistically significant difference in quality between AI-generated and teacher-written IEPs. Most teachers rated the ChatGPT versions as equal to or better than what a human would typically produce — and viewed AI as a tool that could free up time for direct student engagement.

What this means for physics students in Singapore: This is a positive finding about AI as a support tool for educators — not a replacement for teaching, but a genuine relief for the administrative load that eats into teaching time. When educators spend less time on paperwork, students get more of what actually matters: face-to-face, personalised instruction.

At Physics Made Easy, our focus has always been on the quality of direct engagement between Mr. Chew and each student — which is why class sizes are kept small and every session is adapted to the individual learner.

7. 45 Minutes of Daily Recess Reduces Student Stress Hormones by 68%

Researchers compared cortisol levels — measured through hair samples, providing a biological indicator of chronic stress rather than short-term spikes — in fourth-grade students who received either 30 or 45 minutes of daily recess during the school year.

The students who played for 45 minutes per day had 68 percent less cortisol in their hair samples. They were, quite literally, less stressed at a biological level. An exhaustive 2023 review of 50 years of historical records had already concluded that children who spend more time in unstructured outdoor play gain long-term emotional and social benefits — but the 2025 cortisol study puts a biological number on the effect.

What this means for physics students in Singapore: Chronic stress directly impairs the kind of working memory and executive function that physics demands. A student who is perpetually anxious cannot hold multiple variables in mind simultaneously, cannot think flexibly about a problem from different angles, and cannot sustain the metacognitive monitoring that complex problem-solving requires.

Structured downtime — real rest, physical movement, time away from screens and study materials — is not laziness. It is maintenance of the cognitive hardware that learning runs on.

8. Student-Teacher Relationships Are One of the Most Powerful Academic Performance Levers

Two landmark meta-analyses published in 2025 confirm what great teachers have always known — and provide compelling data to back it up.

The first analysed 70 years of studies covering more than 2.6 million K–12 students and found that trusting, supportive student-teacher relationships were linked to higher academic achievement, better behaviour, improved executive function and self-control, and greater feelings of belonging and motivation. The effects were consistent across gender but appeared especially significant for middle and high school students.

The second meta-analysis, covering 40 studies, found that comprehensive social and emotional programmes in grades 1–12 improved overall academic achievement by 8.4 percentile points.

What this means for physics students in Singapore: Physics has a reputation for being a cold, formula-driven subject. But the research is clear: students learn more when they trust their teacher and feel genuinely supported. A student who is afraid to ask a "stupid question" is a student who will never plug the gap in their understanding of electric circuits or force diagrams.

Read our verified student testimonials to see how this plays out in practice — from students who arrived failing and left with distinctions, partly because they finally had a learning environment where confusion was welcome and questions were taken seriously.

9. Practice-Based Teacher Training Produces Better Teachers Than Theory Alone

Researchers compared two approaches to teacher preparation programmes: traditional frameworks emphasising theoretical models, and practice-based approaches involving expert observation and role-playing in simulated classroom environments.

When pre-service teachers rehearsed actual teaching moves with coaches providing real-time feedback — rather than simply reading and discussing theoretical frameworks — they were measurably more proficient when tested in live teaching situations.

The conclusion: in one of the most cognitively demanding and interpersonally complex jobs in the world, practice consistently outperforms theory.

What this means for physics students in Singapore: The same principle applies to learning physics. Reading about wave-particle duality in a textbook is not the same as working through exam questions on it, getting them wrong, discussing why they are wrong, and trying again with that feedback in mind. The practice-based learning loop — attempt, feedback, correction, retry — is the mechanism through which genuine physics understanding develops.

This is why our free resources and study materials are designed to accompany active practice, not replace it — and why our sessions with Mr. Chew focus on working through problems together, not watching solutions being demonstrated.

10. Students Who Write Essays with ChatGPT Remember Almost Nothing They Wrote

This may be the most alarming finding of 2025 — and the most important one for any student tempted to outsource their thinking to AI.

In a 2025 MIT study, college students wrote essays while connected to EEG brain-monitoring equipment. Those who used ChatGPT to write their essays tended to follow the machine's thinking, produced essays that were statistically similar to each other, and showed poorly coordinated brain activity — the neural signature of shallow engagement.

Minutes after completing their essays, only 17 percent of ChatGPT users could recall a single sentence from what they had just written. Students who wrote without AI assistance recalled sentences at rates of 83 to 89 percent.

The researchers found one silver lining: students who wrote their own first drafts and then used AI during revision fared much better. And AI tutors designed to ask probing questions — rather than supply answers — proved genuinely useful as study partners.

What this means for physics students in Singapore: If you use AI to solve your physics problems, you will perform well in practice. You will bomb the exam. This is exactly the pattern we also see with students who copy worked solutions without understanding the reasoning behind each step.

We wrote about this pattern directly in our post on why students know the right answer but still fail physics — and our post on why university physics students struggle to see the big picture expands on what happens when shallow processing accumulates over years.

AI is a powerful revision tool. It is a destructive crutch when it does your thinking for you.

What the 2025 Education Research Tells Us About Learning Physics Well

Taken together, these ten studies paint a remarkably consistent picture. Learning that lasts requires:

A distraction-free environment (no phones)
Active, organised engagement with problems (diagrams, annotations, structured notes)
A pacing that respects cognitive limits (brain breaks, chunked sessions)
Physical, embodied practice (handwriting over typing)
Productive struggle before rescue (hints, not answers)
A relationship built on trust and high expectations
Real practice, not just theory
Your own thinking — not a machine's

Every one of these principles is built into the way Mr. Chew teaches at Physics Made Easy. From the structure of our sessions to the expectations we set for independent problem-solving, the approach is grounded in exactly the kind of evidence these studies represent.

If your child is studying for O-Level, A-Level, or IB Physics in Singapore and you want to experience a teaching approach that is built on how students actually learn — not just what content needs to be covered — book a free trial lesson today.

The science of learning is clear. The question is whether we act on it.

Frequently Asked Questions About Education Research and Physics Learning

Do cell phone bans actually improve physics exam results?

The research strongly suggests yes. The 2025 study of nearly 17,000 students found that phone-free classrooms substantially improve grades — particularly for students who are newer to the material or who are already struggling. For a subject like physics that demands sustained focus, removing smartphones during study sessions is one of the simplest, highest-impact changes a student can make.

How should physics students take notes — by hand or on a laptop?

For physics learning and memory retention, handwriting is significantly more effective than typing. Writing equations, sketching diagrams, and annotating derivations by hand engages deeper cognitive processing. Use digital tools for reference and organisation, but do your active study and problem-solving work on paper.

How long should a physics study session be before taking a break?

Based on the 2025 microbreak research, 10 to 15 minutes of focused study followed by a 90-second break produces better attention and retention than extended uninterrupted sessions. For demanding material like H2 Physics or IB HL Physics, this chunked approach is particularly important.

Is using ChatGPT to study physics helpful or harmful?

It depends entirely on how you use it. Using AI to generate answers to physics problems is harmful — the MIT study showed that students who outsource writing to AI retain almost nothing. Using AI to ask probing questions about a concept you have already attempted, or to explain a step you do not understand after working through a problem yourself, can be genuinely useful.

Why do some physics students understand concepts but still lose marks in exams?

This is one of the most common patterns we see — and the 2025 research on word problems helps explain it. Students who can recognise a concept but cannot organise and apply it under exam conditions are missing the structured problem-solving layer between understanding and execution. Diagrams, annotated working, and systematic organisation of information bridge this gap.

What makes a great physics tutor according to current education research?

The research points to several factors: building a trusting relationship with the student, maintaining high expectations, providing hints rather than answers when students struggle, using practice-based feedback rather than lecturing, and designing sessions that account for cognitive limits like attention span and working memory. You can read about how Mr. Chew applies these principles in our about page and methodology.

How much recess or downtime do physics students actually need?

More than most students currently get. The 2025 cortisol study found that 45 minutes of unstructured daily activity reduced chronic stress hormones by 68 percent compared to 30 minutes. For students in intensive exam preparation, daily physical activity and genuine rest are not luxuries — they are prerequisites for the kind of focused, high-quality thinking that physics demands.

Related reading from the Physics Made Easy Blog:

#Physics#Learning

Latest trends in Education