Do science simulations help you learn? What the evidence says

Can a science simulation help you understand something you have struggled to learn from a lesson? And if it can, do you still need to work with real equipment? Research suggests simulations can help, particularly with understanding scientific concepts. Their value depends on the activity and the learning goal. Choose a virtual laboratory with a specific purpose in mind, and consider where real equipment fits.

Start with what you are trying to learn, and how the activity helps you learn it. Research on conceptual understanding, guidance and physical experiments gives a clearer answer than a general argument for or against screens.

What the evidence supports

A 2012 review by Rutten, van Joolingen and van der Veen found robust evidence that computer simulations can enhance traditional science instruction, particularly laboratory activities. The studies included simulations used to supplement or replace traditional instruction. The findings support using simulations as teaching tools, with their value judged against the activity and its purpose.

Much of the research left an important question open: how much did teacher support, lesson design or the simulation's place in the curriculum influence the results? To understand why a lesson worked, we need to examine what learners were asked to do and the help they received, as well as the software.

What the PhET research says

Banda and Nzabahimana's 2021 review examined 31 experimental or quasi-experimental studies of PhET simulations from the preceding decade. Its focus was conceptual understanding in physics. The authors concluded that PhET simulations can improve that understanding and can be incorporated into active-learning approaches.

The review brought together several studies without calculating a single pooled effect. The studies were not all randomised trials. Its conclusion supports using PhET as part of learning activities, with attention to what learners do once they open a simulation.

For a learner or teacher, this means choosing an activity with a clear idea to investigate. Ask what understanding it is meant to develop. The review concerns understanding physics through PhET. It leaves open questions about percentage increases in marks, other science subjects and other simulation products.

Virtual and physical experiments

Do virtual investigations work better than physical ones? Muilwijk and Lazonder's 2023 meta-analysis compared the two using 35 studies with 3,303 participants. It examined conceptual knowledge in STEM across age groups, including people beyond school age and studies beyond school science.

The analysis found no statistically significant overall advantage for either mode on the conceptual knowledge outcomes examined. That leaves room for differences between the methods and in how well they work for individual learners.

This finding fits alongside the evidence that simulations can improve traditional instruction. The comparisons ask different questions: one compares simulations with ordinary teaching; the other compares two ways of carrying out an investigation.

For school decisions, the range of studies matters. Physics accounted for 23 of the 35 studies. Very few assessed outcomes beyond conceptual knowledge, such as inquiry skills or understanding the nature of science. Questions about equal practical competence, safety skills and long-term retention remain open. Apply the findings to other subjects with care.

What transfers to real equipment?

One study offers a specific example of learning transferring from a simulation to physical apparatus. In Finkelstein and colleagues' 2005 introductory university physics study, students used a circuit simulation that explicitly modelled electron flow. They outperformed the real-equipment group on a conceptual survey and on coordinated tasks involving assembling a real circuit and explaining it.

The result is encouraging within its setting: university students learning about circuits. Whether the learning transfers to other laboratory tasks remains a separate question.

The researchers also made clear that troubleshooting circuits remained a valuable skill. The laboratory's goal was to develop a model of simple circuits; troubleshooting was outside that focus. Name the goal before deciding whether one kind of laboratory work can replace another. Understanding a circuit and learning to troubleshoot it are different goals, and real equipment may serve purposes beyond those tested here.

Combining the two can help

There is evidence for using physical and virtual experiments together. A 2022 systematic review by Wörner, Kuhn and Scheiter covered 42 experimental studies. In most cases, combining the two promoted conceptual understanding more effectively than a single type of experimentation.

Give each activity a clear purpose. Read the 42 studies with their individual comparisons in mind. Combining activities helped in most cases, rather than every case. The benefit depends on more than simply adding a second activity.

The review found no evidence that a particular sequence was superior. Teachers can choose an order suited to the learning task, whether simulations or physical work come first. The findings leave open whether order matters in particular circumstances.

Guidance matters, but its design matters too

Learners can investigate a question with help. Lazonder and Harmsen's 2016 meta-analysis brought together 72 studies of inquiry-based learning and found a positive overall effect of guidance on learning outcomes. It examined support for inquiry generally, rather than estimating PhET's effect.

Evidence about age-related differences was too limited for a firm conclusion. How much help pupils need in each school year remains an open question.

A study by Eckhardt and colleagues shows why the form of support deserves attention. Among 124 eighth-grade pupils using an ecosystem simulation, the highest learning outcomes occurred when pupils received help either with interpreting data or with regulating their learning. The study assessed factual, conceptual and procedural knowledge within that topic.

Combining both forms of support appeared to harm knowledge acquisition in this study. Those pupils also reported the highest cognitive load, meaning the greatest mental demands. The authors suggested excessive load as a possible explanation, though they could not prove it was the cause. The finding concerns these particular forms of support; their combination may work differently in other settings.

How to use the evidence

For students, parents and teachers, start with the intended learning outcome. Ask what concept the activity addresses, what guidance comes with it and what the learner should be able to explain afterwards. These questions can help you choose and discuss an activity, though they are not a tested formula for results.

Keep practical goals in view too. Understanding a concept, troubleshooting equipment and carrying out laboratory tasks are separate goals. Judge the evidence against the goal you have in mind. Simulations deserve a place among learning activities, with attention to teaching, physical experience and what each study actually examined.

To explore further, Learnacy Labs is open in beta and describes a guided Understand, Explore and Prove flow. That is its stated format; the research discussed here did not evaluate Learnacy Labs.