Highlights
Strong evidence alone is often not enough to change professional beliefs.
Teachers naturally interpret new evidence through beliefs they already hold.
Instructional approaches often become symbols of educational values like curiosity, equity, or student agency.
When instructional beliefs become part of our professional identity, changing our minds becomes even harder.
Conceptual change research offers practical ways to make professional learning more effective.
One of the enduring puzzles of teacher education is why strong research evidence so often fails to change teachers’ beliefs about instruction.
Most of us like to believe that if we were shown strong evidence contradicting one of our beliefs, we would simply change our minds. We see ourselves as thoughtful, rational people who are willing to follow the evidence. Yet psychological research suggests that belief revision is much more difficult than this. We do not evaluate new evidence in isolation. We interpret it through the understandings we already possess, sometimes reshaping new ideas so that they fit our existing thinking. The result may be a distorted understanding, or a conclusion quite different from the one intended.
This insight has profound implications for teacher education and professional learning. It suggests that simply presenting compelling evidence is rarely enough to change professional thinking. Genuine conceptual change requires something more.
In this article, we argue that teacher education and professional learning must contend with three progressively deeper layers of resistance to conceptual change. First, our existing beliefs about teaching and learning may interfere with our understanding of the evidence. Second, changing those beliefs becomes harder if we see them as expressing educational values that matter deeply to us. Third, change becomes harder still when we feel those beliefs are part of our identity as educators and connect us to professional communities that share them. Each layer makes genuine conceptual change more difficult than the last.
Recognizing these layers helps explain why evidence sometimes fails to change professional thinking. It also points toward more effective approaches to teacher education and professional learning.
The first layer of challenge: Our need to preserve a coherent understanding
A common assumption is that learning simply involves acquiring new knowledge. We hear a new idea, understand it, and add it to what we already know.
Conceptual change research suggests something quite different.
Human beings naturally seek coherent ways of understanding the world. Rather than treating each new idea as an isolated piece of information, we try to fit it into conceptual frameworks that already make sense to us. If the new information fits comfortably with our existing understanding, learning can be relatively straightforward. But when it conflicts with the way we make sense of the world, we rarely abandon our existing understanding immediately. Instead, we usually reinterpret the new information in ways that preserve as much of our existing understanding as possible.
Psychologists have long recognized this tendency. They distinguish between assimilation—fitting new information into an existing conceptual framework—and accommodation—reorganizing that framework to make room for a genuinely new understanding. Whenever possible, we tend to assimilate rather than accommodate. Assimilation preserves the coherence of what we already believe, whereas accommodation requires us to reorganize our existing understanding—a process that is often considerably more difficult.
One of the classic demonstrations comes from children’s ideas about the Earth. Young children are often taught that the Earth is round, and many can correctly repeat this fact when asked. Yet researchers discovered that children frequently incorporated this new information into their existing flat-Earth conceptions rather than replacing them (Vosniadou & Brewer, 1992). When asked to draw a round Earth, some created flat circular Earths, spherical Earths with everyone living at the North Pole, and even hollow Earths with people living inside.
These are not random mistakes. They are remarkably creative attempts to preserve as much of an existing worldview as possible (i.e., the concept that “down” is a particular direction) while incorporating the new knowledge that the Earth is round.
The implications of this research extend naturally to teacher education and professional development. Whenever new ideas challenge one of our existing conceptual frameworks, our first instinct is usually not to replace that framework but to reinterpret the new information so that it fits what we already believe. Consequently, when new instructional ideas conflict with a teacher’s existing understanding, simply presenting a body of evidence may be insufficient to promote genuine conceptual change. Instead, teachers may assimilate the new information into their existing understanding rather than reorganizing that understanding through accommodation. If our goal is genuine accommodation, we must support teachers in recognizing the limitations of their existing frameworks and in constructing revised ones that are intelligible, plausible, and ultimately more useful.
Consider a teacher who attends a professional learning session on explicit instruction. The presenter discusses the importance of modelling, worked examples, guided practice, frequent checks for understanding, and the gradual release of responsibility. Yet the teacher leaves thinking, “They’re telling us to go back to standing at the front and lecturing.” Like the children in the Earth studies, the teacher has not rejected the new idea. But much of its complexity has been lost. The teacher has recast it as a simpler, more familiar story that better fits what they already believe.
Unfortunately, this is only the first layer of the challenge. In education, our conceptual frameworks are often connected to educational values that matter deeply to us. Once this happens, conceptual change becomes even more difficult.
The second layer of challenge: When our beliefs about instruction become connected to our values
The first layer of challenge arises because teachers interpret new ideas through the lens of what they already believe. A second layer emerges when particular instructional approaches become psychologically associated with broader educational values. At this point, teachers are no longer evaluating instructional approaches solely in terms of whether they make sense. They are also evaluating what those approaches have come to represent.
Many educators are deeply committed to nurturing students’ curiosity, agency, creativity, critical thinking, and independence, while also promoting equity and inclusion. These commitments are among the reasons many of us entered the profession. Difficulties can arise, however, when particular instructional approaches become tightly associated with these valued educational ideals. For example:
Inquiry learning ↔ curiosity and creativity
Student-directed learning ↔ agency and learner independence
Collaborative learning ↔ equity and inclusion
Similarly, other instructional approaches can become associated with values that many educators reject:
Explicit instruction ↔ teacher control rather than student empowerment
Knowledge-rich teaching ↔ memorization rather than critical thinking
The problem is not the values themselves, which most educators endorse. Rather, the problem is that certain instructional approaches can come to symbolize those values. When a strong association is forged, questioning an instructional approach can feel like questioning the underlying value itself.
Consider a teacher who has come to associate inquiry learning with curiosity, creativity and student agency. When presented with evidence that explicit instruction is often beneficial for novice learners, the teacher may initially respond with skepticism, because accepting it seems to conflict with their deeply held values. They might assume the evidence is wrong. Or they might make presumptions about the motivations of the presenter (e.g., “This presenter doesn’t understand the importance of fostering student agency.”) This makes conceptual change considerably more difficult because the discussion is no longer experienced as being about the evidence alone. It is now seen as also being about educational values.
Many of these associations do not arise accidentally. Sometimes they are explicitly taught. Teacher education programs, professional learning sessions, policy documents, curriculum resources, social media, and professional conversations may present particular instructional approaches as natural expressions of important educational values. Teachers may be told, for example, that “student-centered approaches empower learners,” or “teacher-led instruction is authoritarian / outdated.” As a result, evidence supporting or questioning a particular instructional approach may come to be interpreted as supporting or questioning the educational values with which that approach has become associated.
An evidence-informed perspective invites a different way of thinking about instructional approaches. Rather than viewing a technique as inherently good or bad, it asks under what conditions it is most likely to support learning. Explicit instruction, inquiry, collaborative learning, discussion, worked examples, and problem solving all have important roles to play, but their effectiveness depends on factors such as learners’ prior knowledge, instructional goals, task complexity, and the support that is provided. This more nuanced understanding becomes difficult to develop when particular instructional approaches come to symbolize broader educational values. Curiosity, creativity, equity, critical thinking, and learner independence are not the products of any single pedagogy. Rather, they are long-term educational goals that are typically fostered through the thoughtful combination of many different instructional approaches over time.
The third layer of challenge: When instructional beliefs become part of our identity
There is one further layer of challenge. Over time, value-infused instructional beliefs can become part of our professional identities and the communities to which we belong. We begin to think of ourselves as inquiry teachers, progressive educators, explicit instruction advocates, constructivists, traditionalists, or members of countless other educational communities.
To be fair, these communities are enormously valuable. They provide support, professional learning, and opportunities to refine our thinking. But they also create subtle pressures.
When a belief becomes part of our identity, changing our minds can feel like distancing ourselves from the people with whom we most identify. Without consciously realizing it, we may begin evaluating evidence using a different question:
What will people like me think if I accept this conclusion?
This helps explain why educational debates can become so polarized. Once instructional approaches become connected to professional identity and group membership, disagreements are no longer simply about instructional methods or research findings. They also become questions of belonging, loyalty, and the kind of educator we see ourselves as being. Under these conditions, evidence that challenges an instructional belief can feel like a challenge to our identity itself.
None of us is immune. An educator who thinks of themselves as an inquiry teacher may become dismissive of research supporting explicit instruction. Likewise, someone who identifies strongly with explicit instruction may become equally dismissive of situations in which inquiry, collaboration, or discovery genuinely contribute to learning. In both cases, the disagreement is no longer simply about evidence. It is also about protecting an identity.
The language of social media discussions often makes these identity dynamics visible. Like-minded thinkers may be described as “allies,” while those who disagree become “opponents,” “critics,” or “adversaries.” Such language frames disagreement less as a collective effort to understand evidence and more as a conflict between competing sides. Under these conditions, changing one’s position may feel less like revising a belief than disappointing one’s allies or conceding ground to an opposing camp.
What this means for evidence-informed teacher educators
If conceptual change is as psychologically complex as the research suggests, then presenting evidence alone is unlikely to be enough. Educators interpret professional learning through the lens of their existing beliefs. Some of those beliefs have become intertwined with their educational values and, over time, even with their professional identities. Under these conditions, simply presenting strong evidence may not be sufficient to bring about conceptual change.
How should teacher educators respond to this situation? The conceptual change literature offers a fairly coherent answer (see Posner et al., 1982; Gregoire, 2003; Sinatra & Pintrich, 2003). Here are some suggestions:
Begin by eliciting existing beliefs. Rather than assuming participants arrive as blank slates, invite them to explain how they currently think about teaching and learning. Asking educators to predict outcomes, explain why they use particular instructional approaches, or describe what they believe makes learning effective brings their existing conceptions into the open. Beliefs that remain implicit are difficult to examine or revise.
Treat beliefs respectfully. Most teachers’ beliefs developed for understandable reasons. They reflect years of classroom experience, successful lessons, and genuine concern for students. Dismissing some beliefs as simply “wrong” is unlikely to encourage reflection. Conceptual change is more likely when people feel their existing thinking has been understood before being challenged.
Separate instructional methods from educational values. This is a key step. Problems can arise when teachers’ beliefs about particular instructional approaches become psychologically intertwined with important educational values. The challenge for teacher educators is disentangle values from the methods with which they have become associated. The question is not whether values like curiosity, equity, creativity, or learner agency matter—they do. Openly affirm them. Then turn to the evidence and examine which instructional approaches are most likely to achieve these goals in particular instructional contexts, and why.
Teach instructional approaches as context-dependent tools rather than educational philosophies. Few instructional approaches are inherently good or bad. Explicit instruction, inquiry, collaborative learning, discussion, problem solving, and worked examples all have important roles to play. The critical question is not Which approach is best? but Under what conditions is this approach most likely to support learning? Framing instructional approaches in this way encourages conditional rather than ideological thinking.
Provide coherent alternatives rather than isolated findings. Research findings are more persuasive when they fit within a broader explanation of how learning works. Rather than simply presenting individual studies, help educators understand the underlying principles that connect ideas such as prior knowledge, working memory, cognitive load, retrieval practice, and explicit instruction into a coherent model of learning.
Create opportunities for sustained reflection. Conceptual change rarely occurs during a single workshop. Teachers need repeated opportunities to compare their existing beliefs with new evidence, apply alternative explanations to authentic classroom situations, discuss their thinking with colleagues, and revisit their conclusions over time.
Finally, teacher educators should recognize that changing one’s mind is not a sign of professional weakness—it is one of the hallmarks of professional expertise. Professional learning should cultivate intellectual humility by creating environments in which educators can question long-held assumptions without feeling that they are abandoning their values or identities.
What this means for evidence-informed learners
Conceptual change research also has important implications for all of us, as learners.
When we encounter research that challenges our preferred instructional approaches, our first response should not simply be to question the research, or ask whether it is persuasive. We should also ask ourselves why we are reacting the way we are.
Am I rejecting this evidence because it is genuinely unconvincing? Or because it conflicts with beliefs I have held for many years? Has this instructional approach become connected, in my mind, to values that matter deeply to me? Has it become part of my professional identity?
These are not easy questions to ask ourselves. Yet they may be among the most important habits of evidence-informed practice.
None of this implies that teachers should abandon their professional judgement or accept every new claim that appears in the research literature. Healthy skepticism is an essential part of evidence-informed practice. The challenge is to ensure that our skepticism is directed toward the quality of the evidence rather than unconsciously shaped by our prior beliefs, values, or identities.
One practical habit is to run periodic “self-checks” in which we compare our intuitions to the strongest available evidence. Whenever we encounter a new instructional idea—whether in a professional development session, a book, a podcast, or on social media—we can ask: What does the best available research actually say? Likewise, when we encounter research that surprises us or instinctively feels wrong, that is often a signal to investigate further rather than dismiss it outright.
AI tools such as the Evidence Checker can support this process by helping educators compare their initial impressions with current evidence from the research literature. The goal is not to outsource professional judgement to an AI tool or to accept every conclusion uncritically. Rather, it is to create a deliberate pause between our immediate reactions and our final conclusions, giving us an opportunity to reflect on whether our prior beliefs, values, or identities may be influencing how we interpret the evidence.
Finally, we should think carefully about the role of our professional communities. These communities are invaluable sources of support, collaboration, and professional growth. But if all they do is reinforce what we already believe, they are more of a hindrance than help. Professional communities should challenge our thinking, not simply reinforce it. The healthiest professional cultures are those in which educators can question long-held assumptions, change their minds without losing the respect of their colleagues, and see intellectual humility as a sign of professional strength rather than weakness.
Ultimately, becoming evidence-informed is not simply about learning more research. It is also about learning more about ourselves. Recognizing the psychological forces that influence how we interpret evidence may be one of the most important steps toward genuine professional growth.
References
Gregoire, M. (2003). Is it a challenge or a threat? A dual-process model of teachers’ cognition and appraisal processes during conceptual change. Educational Psychology Review, 15, 147–179. https://doi.org/10.1023/A:1023477131081
Posner, G. J., Strike, K. A., Hewson, P. W., & Gertzog, W. A. (1982). Accommodation of a scientific conception: Toward a theory of conceptual change. Science Education, 66(2), 211–227.
Sinatra, G. M., & Pintrich, P. R. (Eds.). (2003). Intentional Conceptual Change. Mahwah, NJ: Erlbaum.
Vosniadou, S., & Brewer, W. F. (1992). Mental Models of the Earth: A Study of Conceptual Change in Childhood. Cognitive Psychology, 24(4), 535–585.
Love these reflections!
I have also seen and heard: “Oh, I already do that,” but the instructional practices in the classroom don’t always match the belief.
Take for instance Building Thinking Classrooms. There are some elements within the framework that one could argue are explicit instruction. Like, his emphasis on “thin slicing” (I.e., comparable to chunking steps), “hints”provided as students need them (scaffolding and prompting), teacher circulating to observe student work at boards (monitoring understanding), lesson consolidation (bringing lesson back to the important mathematical principle), formative assessment (using evidence to determine what needs reteaching) and practice through increasingly complex problems (some could argue gradual release).
The central difference is the order in which instruction and problem-solving occur. It also expressly goes against using worked examples at the start of a lesson, and outright refutes teacher-led consolidation, rather arguing that students should instead explain each others’ thinking at the close of a lesson.
So while BTC is explicit about classroom routines, task sequencing and facilitation, it is not generally explicit about initial knowledge acquisition. However, teachers and instructional coaches will challenge my take. They argue that what they are doing is very much explicit teaching. Furthermore, as I observe classrooms where BTC has taken complete hold, I rarely see it done well. It is highly complex, requires the teacher to have a level of content knowledge that, these days in education, is often hard to find, and takes months to get students to attend to the routines.
This is another layer of the research challenging the belief which I have encountered as I attempt to support teachers on improving their practice.
This is really helpful. Thank you.