Physics education has a well-documented problem. Students can pass exams — reproduce derivations, solve standard problems — while holding deeply incorrect conceptual models. A student who can calculate the trajectory of a projectile may still intuitively believe that heavier objects fall faster. A student who can solve the Schrödinger equation may not understand what it means for a particle to be in superposition. This gap between procedural competence and conceptual understanding is the central challenge of physics pedagogy.
Interactive simulations address this gap directly. They make abstract phenomena concrete, allow rapid experimentation, and create the conditions for what educational researchers call "productive failure" — the learning that comes from making and recognising mistakes in a safe environment.
The Research: What We Know About Simulation-Based Learning
The University of Colorado's PhET Interactive Simulations project, led by Nobel laureate Carl Wieman, has produced some of the most rigorous evidence for simulation-based learning. Their research consistently shows that well-designed interactive simulations produce greater conceptual learning gains than traditional instruction alone — particularly for topics involving invisible or abstract phenomena like electric fields, quantum mechanics, and molecular dynamics.
The key finding is that the interactivity is essential. Passive videos of simulations produce significantly smaller gains than interactive simulations where students control parameters. The act of predicting, adjusting, and observing — closing the experiment-result loop — is what drives learning.
Productive Failure: Learning from Getting It Wrong
Educational researcher Manu Kapur developed the Productive Failure framework through a series of studies in Singapore secondary schools. Students who attempted a novel problem before receiving instruction outperformed students who received instruction first, even though the first group's initial attempts were wrong or incomplete.
Simulations create ideal conditions for productive failure. A student who tries to make a stable orbit in the N-body simulation and watches their planet spiral into the star learns something about the relationship between orbital speed, altitude, and stability that no diagram can convey. When they subsequently learn Kepler's laws, those laws explain the failure they experienced rather than describing something abstract.
Concrete Before Abstract: Enactive Learning
Educational psychologist Jerome Bruner identified three modes of representation in learning: enactive (doing), iconic (seeing), and symbolic (abstract notation). Effective physics instruction should move through these stages — first doing, then seeing, then symbolising. Traditional physics courses often present the symbolic stage first (equations and derivations) with only brief enactive exposure in laboratory sessions.
Simulations compress the enactive and iconic stages and make them available outside the laboratory. A student who has spent twenty minutes playing with the fluid simulation — dragging obstacles, watching vortices form, noting how pressure backs up behind a barrier — has a concrete, embodied understanding of fluid dynamics that primes them to make sense of the Navier-Stokes equations when they see them.
See fluid dynamics made tangible
Try our Fluid Simulation as a pre-lecture activity before introducing fluid mechanics. Students who have explored it arrive with concrete intuitions about pressure, flow separation, and turbulence that dramatically improve lecture comprehension.
Practical Tips for Teachers
Before the Lesson
Assign a simulation as a pre-class exploration task. Give students a small number of specific questions to investigate: "What happens to the N-body system's total energy as you decrease the timestep? What does this tell you about numerical integration?" Require a short written prediction before they start and a comparison with the result after. This predict-observe-explain cycle is the most effective structure for simulation-based learning.
During the Lesson
Use the simulation as a shared demonstration with deliberate pause points. Show a behaviour, stop, ask for explanations, then reveal the physical mechanism. The simulation makes the phenomenon visible; the lesson makes it explicable. For topics like chaos theory, the simulation can show something genuinely surprising — the divergence of two nearly identical double pendulums — that creates cognitive dissonance and motivates the mathematical explanation.
After introducing a concept, challenge students to find simulation parameters that break the expected behaviour. "Find settings where the SPH fluid becomes unstable." "Find a double pendulum starting angle where the motion remains periodic." The search for edge cases develops deep understanding of the model's assumptions and limits.
For Assessment and Investigation
Simulations provide excellent platforms for data-collection investigations. Students can:
- Measure the largest Lyapunov exponent of the double pendulum by tracking divergence time vs initial separation
- Verify the conservation of angular momentum in the orbital simulation by measuring before and after a simulated gravitational slingshot
- Map the Mandelbrot set's period-doubling cascade and measure the Feigenbaum constant
- Measure how SPH fluid's pressure wave speed varies with density and compare to the theoretical speed of sound
These are genuine physics investigations with real data and expected theoretical outcomes. They develop experimental skills — data collection, uncertainty estimation, model fitting — in a controlled environment where the "experiment" can be repeated exactly.
Addressing Common Concerns
Won't students just play without learning? The research shows that unguided exploration of simulations produces smaller gains than guided exploration with specific questions. Structure matters. The simulate-then-lecture sequence consistently outperforms lecture-then-simulate for conceptual retention.
Are simulations a substitute for real experiments? No — and they work best as complements, not replacements. Real experiments teach procedural skills, measurement uncertainty, and the frustration of equipment that doesn't behave. Simulations teach conceptual models and allow rapid parameter sweeps that no physical apparatus could match. Both are essential.
What about students without reliable internet access? All MySimulator simulations run entirely in the browser with no server round-trips after the initial page load. They work on tablets, older laptops, and low-bandwidth connections. The page loads once; after that, the simulation runs locally.
Start with our curated student list
See our Top 10 Simulations for Physics Students for a curriculum-aligned selection with notes on which university modules each simulation supports — and try all of them from the MySimulator home page.