Conceptual change is the process by which a person fundamentally revises a deeply held belief or mental model, replacing it with a new understanding that better accounts for the evidence. It is not the same as simply adding new facts to what you already know. When a child learns that the Earth is a sphere despite every daily experience suggesting the ground is flat, or when a physics student genuinely grasps that a book resting on a table pushes down on the table and the table pushes back up on the book, something more radical than memorization has to happen. The research on how this process works, why it so often fails, and what makes it succeed has grown into one of the most active areas in the learning sciences over the past four decades.
Where the Idea Came From
The study of conceptual change in education took shape in the 1980s and 1990s, drawing heavily from Thomas Kuhn’s philosophy of science. Kuhn’s argument that scientific progress happens not through steady accumulation but through disruptive “paradigm shifts” inspired researchers to ask whether individual learners go through something similar. His work directly influenced the emergence of conceptual change research as a distinct field within science education, alongside constructivism and cultural studies of science.1PubMed Central. Thomas Kuhn and Science Education: Learning from the Past and the Importance of History and Philosophy of Science The parallel is not perfect, but the core insight holds: people do not just passively receive correct information and update their beliefs. They hold existing frameworks that actively resist replacement.
Why Your First Ideas About the World Are So Stubborn
From a very early age, children build intuitive theories about how the physical and biological world works. These are not random guesses. They are coherent frameworks that organize everyday experience into something predictable. Children develop intuitive ideas about what counts as alive versus not alive, about how objects move and fall, and about what other people are thinking. Research on these “naive theories” shows that they are domain-specific: a child’s understanding of biology operates somewhat independently from their understanding of physics or psychology.2PubMed Central. Domains and naïve theories
This matters for conceptual change because it means a person’s wrong idea about, say, how electric circuits work is not a loose misconception floating in isolation. It is embedded in a broader framework of physical intuitions that have been reinforced by years of everyday experience. Correcting one piece often means destabilizing a whole interconnected web of beliefs, which is exactly why people resist it. The everyday experience that grounds these intuitive theories is not wrong, either. It is just incomplete. Gravity does pull things down. Heavier things do feel like they should fall faster. The ground does look flat. The intuitive theory works well enough for navigating daily life, so the motivation to abandon it for some abstract scientific model can be low.
Work with children on the autism spectrum has helped sharpen this picture. Studies comparing children with ASD to typically developing children found that naive reasoning about physics and biology was equivalent across both groups, even though children with ASD struggled more with tracking false beliefs in other people.3PubMed. Naïve Theories of Biology, Physics, and Psychology in Children with ASD That dissociation supports the idea that intuitive theories about the physical world are genuinely separate cognitive structures from social reasoning, and both populations build them in similar ways.
How Children Invent Compromise Models
One of the most vivid demonstrations of conceptual change in action comes from research on how children understand the shape of the Earth. In a study of first-, third-, and fifth-graders, children were asked questions about the Earth’s shape. Their answers looked inconsistent on the surface: many said the Earth is round but also said it has an edge from which people could fall off. That sounds contradictory, but it stopped looking contradictory once researchers realized the children were not confused. They were working with coherent mental models that simply were not the scientific one.4Cognitive Psychology. Mental models of the earth: A study of conceptual change in childhood
Five distinct models emerged from the data. Some children pictured the Earth as rectangular or disc-shaped. Others invented “synthetic” models: a hollow sphere with people living on a flat surface inside, or a flattened sphere. These synthetic models are fascinating because they represent the child’s attempt to reconcile two conflicting inputs. They know adults keep saying the Earth is round, and they can see it looks flat. So they invent a compromise: maybe it is round on the outside but flat on the inside where people stand. Children only arrived at a genuine spherical-Earth understanding once the underlying assumptions that generated their initial models had been fundamentally reinterpreted. Simply being told the correct answer was not enough. The presuppositions had to shift first.
This pattern of synthetic models, where learners create hybrids of old and new ideas, shows up well beyond Earth science. It is one of the hallmarks of conceptual change in progress. A student who has been taught Newton’s third law but still privately believes that a truck exerts more force on a car in a collision than the car exerts on the truck is holding a synthetic model. They can parrot the correct answer while retaining the old framework underneath.
What Brain Imaging Reveals
One of the most provocative findings in conceptual change research is that scientific experts do not seem to erase their old misconceptions. They override them. Brain imaging studies have tested this by showing physics experts and novices statements that are either scientifically correct or consistent with common misconceptions. When experts evaluated incorrect statements, they showed significantly greater activation in brain regions associated with inhibitory control, specifically the right ventrolateral prefrontal cortex and the left dorsolateral prefrontal cortex.5Trends in Neuroscience and Education. Is inhibition involved in overcoming a common physics misconception in mechanics? Novices did not show this same pattern. The implication is striking: the experts’ old, incorrect intuitions were still there in their brains, and they had to actively suppress them to give the right answer.
This pattern has been replicated across different physics topics. In a study on electricity, experts evaluating circuits that depicted a common misconception (a bulb lighting up with only one wire connecting it to a battery) also showed greater activation in inhibition-related brain areas compared to novices.6Mind, Brain, and Education. Differences in Brain Activation Between Novices and Experts in Science During a Task Involving a Common Misconception in Electricity The misconception appeared to remain encoded in the experts’ neural networks, requiring active suppression every time.
A broader neuroimaging study of scientists with PhDs in physics found that when they encountered naive ideas that contradicted scientific knowledge, they showed increased activation in the left inferior frontal gyrus, bilateral superior frontal gyrus, bilateral anterior cingulate cortex, bilateral cerebellum, and middle frontal gyrus.7npj Science of Learning. An fMRI study of scientists with a Ph.D. in physics confronted with naive ideas in science Several of these regions are involved in conflict detection and cognitive control. Even for highly trained scientists, encountering a statement that aligns with an old intuitive model triggers a measurable neural conflict that requires effortful resolution.
This reframes what “learning the right answer” actually means at a biological level. Conceptual change may not involve deleting the old idea. It may instead involve building a parallel, correct understanding and then developing the capacity to consistently choose it over the old one. The old idea remains a kind of cognitive ghost. For the expert, inhibiting it has become automatic enough to feel effortless, but the brain scans tell a different story.
How People React to Evidence That Contradicts Their Beliefs
If old ideas persist and require active override, you might wonder what happens when a learner encounters evidence that directly challenges their existing belief. The short answer is: many things, and only some of them are productive. Research on how students respond to anomalous data identified a wide taxonomy of reactions. In one study, younger students were presented with information about the extinction of the dinosaurs and asked to evaluate evidence. Researchers found 24 distinct categories of reasons students gave for accepting or rejecting the evidence, which mapped onto a taxonomy of eight types of responses to anomalous data.8Learning and Instruction. Responses to anomalous data on controversial topics and theory change
These responses range from fully accepting the new data and revising one’s theory, down to outright rejection. In between lie several more subtle strategies: reinterpreting the data to fit the existing belief, holding the new information in a separate mental category without letting it affect the core theory, or acknowledging the data but deciding it is not relevant to their current understanding. The overwhelming majority of students in that study favored one particular explanation for the dinosaurs’ extinction, suggesting that even when presented with multiple lines of evidence, learners tend to latch onto a single narrative that feels most coherent to them. The tendency to protect an existing framework is not obstinacy. It is how human cognition handles complexity.
Teaching Strategies That Push Past Resistance
Given how stubbornly old ideas persist, educators have tested a variety of approaches specifically designed to trigger genuine conceptual change rather than surface-level memorization. The evidence favors a few strategies in particular.
Refutation Texts
A refutation text is a passage that explicitly states a common misconception, directly contradicts it, and then explains the correct concept. This structure forces the reader into a kind of cognitive confrontation. You cannot passively skim it the way you might skim a standard textbook explanation, because the text names the wrong idea you probably hold and tells you why it is wrong before offering the replacement. A meta-analysis pooling 44 independent comparisons across nearly 3,900 participants found that refutation text produces a moderate positive effect on learning compared to other instructional approaches, and this effect was consistent across a wide range of topics and contexts.9PubMed Central. Refutation Text Facilitates Learning: a Meta-Analysis of Between-Subjects Experiments
An earlier study comparing refutation texts to standard explanatory texts in biology found something surprising: both types of text led to conceptual change in a large proportion of students.10Science Education. Investigating the relationship between refutational text and conceptual change That result complicates the picture. It suggests that under certain conditions, even a well-written standard explanation can shift beliefs. But across the broader evidence base, the refutation structure consistently outperforms standard approaches, probably because it makes the conflict between old and new ideas impossible to ignore.
Bridging Analogies
Another effective approach uses what are called bridging analogies: a sequence of examples that starts from a situation where the student’s intuition is correct and gradually moves toward a situation where that intuition breaks down. In physics, for instance, most students intuitively accept that a spring pushes back when you compress it. They are less willing to accept that a rigid table pushes back up on a book resting on it. A bridging analogy sequence connects these cases through intermediate examples (a flexible board, a piece of foam) that make the transition feel natural.
Research on bridging analogies in teaching Newton’s third law found that they were effective at engaging students with the idea of action-reaction forces, and there was evidence that for some students, bridging analogies produced greater conceptual change than straightforward didactic instruction.11International Journal of Science Education. Encouraging conceptual change: The use of bridging analogies in the teaching of action-reaction forces and the ‘at rest’ condition in physics Students were able to map each analogy to the target concept and use the sequence to refine their causal explanations. The approach also seemed to support metacognitive skills: students became more aware of how their own understanding was shifting.
The effectiveness of analogies hinges on a few conditions. The examples must be understandable and believable to the student, not just to the teacher. Even when an example is compelling, students may not spontaneously see it as analogous to the target problem, so the analogy relationship often needs to be drawn out explicitly. And qualitative, visualizable models that give mechanistic explanations tend to work better than abstract ones.12Journal of Research in Science Teaching. Using examples and analogies to remediate misconceptions in physics: Factors influencing conceptual change The analogy is doing real cognitive work here: it gives the learner a new mental model to latch onto, not just a correct statement to memorize.
Interactive Simulations
Computer-based simulations, like PhET simulations for physics, offer another route. They let students manipulate variables and see the consequences in real time, which can directly confront misconceptions in ways that static text cannot. Reviews of this literature have found that interactive simulations are effective at teaching modern physics concepts and enhancing critical thinking and problem-solving skills.
Motivation Is Not a Side Issue
For years, conceptual change was studied primarily as a “cold” cognitive process: a matter of logic, evidence evaluation, and the structure of knowledge. That framing turned out to be incomplete. Research has identified at least four motivational constructs that mediate whether conceptual change actually happens: the goals a learner holds, how much they value the subject, their self-efficacy (belief in their own ability to learn the material), and their sense of control over the learning process.13Review of Educational Research. Beyond Cold Conceptual Change: The Role of Motivational Beliefs and Classroom Contextual Factors in the Process of Conceptual Change
A learner who does not believe they can master the material, or who does not see why it matters, is far less likely to do the hard cognitive work that genuine revision demands. This goes beyond just “trying harder.” Motivation shapes whether a learner engages deeply enough with conflicting evidence to experience the kind of productive discomfort that precedes real change. A longitudinal case study illustrated this nicely: a student who initially lacked confidence gained it through belonging to a supportive small group, which led to deeper engagement with the material, which in turn produced high levels of conceptual change. The student also shifted from performance-oriented goals (getting the grade) toward mastery-oriented goals (genuinely understanding), and reported increased interest in science overall.14Disciplinary and Interdisciplinary Science Education Research. ‘Even though it might take me a while, in the end, I understand it’: a longitudinal case study of interactions between a conceptual change strategy and student motivation, interest and confidence
The practical implication is that teachers aiming for conceptual change need to attend to more than just the content. The classroom environment, group dynamics, and how students feel about themselves as learners all feed into whether the cognitive conditions for revision are met.
Why Arguing With Peers Helps, But Not Right Away
Collaborative argumentation, where students discuss and debate scientific ideas together, has emerged as one of the more effective social approaches to conceptual change. But the timeline is counterintuitive. Research has found that collaborative argumentation has a delayed but long-lasting effect: the improvement in understanding often does not show up immediately after the discussion but becomes significant during a delay period afterward.15PubMed Central. Long-Lasting Conceptual Change in Science Education: The Role of U-shaped Pattern of Argumentative Dialogue in Collaborative Argumentation
The pattern of dialogue that predicted long-lasting change followed a specific shape: groups that frequently engaged in deliberative argumentation (carefully considering evidence and alternatives) and co-consensual construction (building shared understanding together), while rarely falling into disputative argumentation (trying to win rather than understand), showed the greatest gains over time. There is compelling evidence that peer discussion triggers a kind of metacognitive process: a private mental settling of differences that continues after the conversation ends.16Learning, Culture and Social Interaction. The intertwined effect of collaborative argumentation and whole-class talk on the process of scientific concept learning: A case study This matches everyday experience. You sometimes leave a good argument feeling more confused than when you started, but a day later you realize the other person had a point and your thinking has quietly shifted.
The delayed nature of the effect has practical consequences for assessment. If a teacher tests students immediately after a group discussion, the results may understate how much learning actually took place. The real gains may only become visible on a follow-up test days or weeks later.
What Eye Tracking Tells Us About Attention During Learning
Newer research has started using eye-tracking technology to observe conceptual change as it happens, watching where learners direct their attention and how that attention shifts during the process. In one study, students who successfully achieved conceptual change while learning from a computer-based environment showed more fixation points in areas of interest and significantly more bidirectional connections between those areas. Students who failed to revise their misconceptions made fewer fixations on the relevant areas and showed more unidirectional scanning patterns, as though they were reading through the material without deeply connecting its parts.17Education and Information Technologies. Unfolding the cognitive process underlying computer-based scientific conceptual change with eye tracker: Behavioral performance and sequential analysis of attention
An eye-tracking study on projectile motion problems found something that dovetails with the brain imaging research on inhibition. When students encountered problems designed to trigger a common misconception (that a longer range means a longer flight time), their eyes were initially drawn to the misconception-related feature, specifically the longer range. Those who overcame the misconception then shifted attention away from that salient but misleading feature and invested additional attention in processing the correct information. Reaction times were longer when students had to overcome the misconception, consistent with the idea that inhibitory control was engaged.18Physical Review Physics Education Research. Role of inhibition in overcoming interferences of misconception under similar feature saliency: An eye-tracking study of the projectile motion problem This kind of evidence suggests that successful learners are not just thinking differently from unsuccessful ones; they are looking differently, too. Their visual attention reflects the ongoing cognitive battle between old intuitions and new understanding.
Eye-tracking also opens up possibilities for real-time intervention. If software can detect that a student’s gaze pattern resembles the unsuccessful pattern, in theory it could redirect attention before the misconception solidifies. That application is still largely experimental, but it represents a shift from studying conceptual change after the fact to potentially supporting it as it unfolds.