Why Mythbusting Fails: A Guide to Influencing Education With Science

By Michael Pershan and Benjamin Riley. We wrote it together, and Ben posted it over at the Deans for Impact blog. I think there are a lot more educators than scientists who read my blog, but pretty much everything but the details apply to an educator interested in helping scientists understand classroom teaching better. Or your daily life, political persuasion, arguments with a friend, etc. – MP

The persistence of neuromyths

“If it disagrees with experiment, it’s wrong,” physicist Richard Feynman said. “In that simple statement is the key to science.”

By this measure, the learning-styles hypothesis has failed too many times to count. Experiment after experiment has shown that matching the form of instruction to a student’s preferred “style” of learning – such as auditory or visual – does not improve a student’s understanding. As a result, the vast majority of cognitive scientists are certain that learning styles have been debunked.

And yet many educators still believe learning styles are important. What gives? Why don’t they trust the science?

One possibility is that educators are simply unaware of the research undermining learning styles’ usefulness. If this is the problem, then there is a simple solution: just spread the word!

To that end, recently a group of 30 learning scientists – including Steven Pinker, Hal Pashler and others – published a letter to inform teachers that learning styles is a “neuromyth” that “create[s] a false impression of individuals’ abilities, leading to expectations and excuses that are detrimental to learning in general, which is a cost in the long term.”

But what if educators are presented with this information…and they still don’t change their minds?

Take Terry J., a lifelong public educator from Canada. Terry read the scientists’ letter, but it failed to convince her to change her practice. “I know that I am a visual learner,” she said. “There is research that supports the idea that learning styles matter, and there is research that says learning styles are bunk.”

Perhaps Terry just “hates scientists.” But this plainly isn’t true – Terry is very interested in science and research that can inform education practice. In fact, she helps design professional development for her province, and she and her team use educational research to plan their workshops. She even consults with a research foundation located just down the hall from where she works.

Terry’s seen the reports, but – in a striking bit of symmetry – she believes that when it comes to learning styles, it’s the research community that is uninformed. The perspective of the letter-writing neuroscientists, she said, is just “based on different assumptions and interpretations.”

The scientists who signed this letter would of course find much to critique in Terry’s views. But if their letter was intended to persuade anyone, surely Terry sits squarely within the target market. She is a public servant, protected from profit motives, with a longstanding interest in applying educational research in practice.

If these scientists can’t persuade Terry, who can they?

The challenge of cultural cognition

Terry’s resistance to scientific authority might seem bizarre – certainly it will cause some learning scientists to shudder. Yet this sort of resistance is not unique to education. In fact, it’s easy to find examples of resistance to scientific insights even in fields that are thought to be highly scientific, such as medicine. And, even more paradoxically, this resistance to scientific evidence can coexist with a strong trust in science and scientists.

How can this be? The answer may lie with what some researchers describe as “cultural cognition.”

Cultural cognition describes how we interpret certain facts and evidence through the lens of our existing values. Usually, we accept scientific claims as true because, overall, most of us trust science and scientists. But – in rare but notable cases – our stance on a scientific matter comes to take on a larger, much more personal meaning. Beliefs about science can become entangled with our self-identities, even if they didn’t start out that way.

Take climate change. Despite mounting evidence and a clear scientific consensus on the relationship between human activities and the rise of global temperatures, beliefs about the cause of global warming are growing more polarized in the U.S., rather than less. From the cultural-cognition perspective, this is largely because beliefs about global warming have become statements about who we are – namely, whether we self-identify as environment-protecting liberals or industry-defending conservatives. This means that increasingly Americans not only disagree on the level of risk posed by global warming, but whether there is even a scientific consensus.

This outcome was not inevitable. But many climate-science advocates emphasize the (perceived) ignorance or anti-science attitudes of those who don’t understand global warming or its causes. In doing so, these advocates effectively insult precisely the people they wished to persuade. The tragic result? The normal bipartisan trust in science has become “polluted” on climate change, in part because of communication strategies that have antagonized existing values, and activated cognitive defenses.

The perils of threatening teacher autonomy

So now let’s return to education and learning styles. What we want to suggest – tentatively and with caveats – is that we run the risk of polluting the environment on learning sciences in the same way people have polluted the climate-change communication environment.

In particular, we worry that some researchers do not fully appreciate the importance that educators place in their own autonomy. Teachers are the ultimate deciders of what takes place in their classrooms, an autonomy that provides them with a major source of professional satisfaction. Teachers may not receive high wages or status, but they do receive tremendous psychic rewards when students appear to learn as a result of their decisions. And educators possess a great deal of (reasonable) sensitivity in protecting this autonomy, as there is a long history of “outsiders” seeking to tinker with what happens within districts, schools and classrooms.

So the scientific consensus that learning styles do not exist will become irrelevant if educators come to see their beliefs about learning styles as critical to their professional autonomy. And one way to heighten the risk of that happening is through talk of what science demands teachers do or believe.

Now, the caveats. We don’t have any direct evidence that teachers currently believe in learning styles because they see it as necessary to establish their professional identities. Nor are we familiar with any evidence tracking changes in educator beliefs about learning styles over time. For all we know, educators who read the recent letter from 30 scientists denouncing learning styles as a “neuromyth” are busy reshaping their beliefs and changing their practice.

But suppose that our analysis is correct. This poses a very delicate dilemma for advocates of learning science (ourselves included). After all, scientific evidence should have weight, and if educator autonomy extends to believing in myths, well, that’s undeserved autonomy.

Is there no hope for change?

Teaching learning science to educators

Let’s return to Terry J. She read the letter by leading scientists denying the existence of learning styles, but nevertheless continued to believe in the debunked hypothesis. Attempts to persuade Terry to abandon this neuromyth may backfire if they emphasize her obligation to accept the burden of scientific evidence, especially if Terry sees this as a threat to her professional autonomy. This is a vexing challenge.

There is a way forward. We need more teaching – and less preaching – to influence the beliefs of educators such as Terry. To achieve this, advocates of learning science should borrow from the playbooks of good science teachers. These teachers do not prioritize getting students to reject their existing beliefs, but instead seek to foster new scientific knowledge in their students. They replace scientific misconceptions, rather than debunk them.

But how should learning science be taught to teachers? We urge learning-science advocates to ask three questions before attempting to influence teachers.

1. What do educators already believe about how learning takes place, and why?

A bedrock principle of cognitive science is that we learn new ideas by reference to what we already know. Effective teachers are eager to understand their students’ existing beliefs so that they (the teachers) can use prior beliefs and understanding to develop new knowledge.

We think advocates of learning science should be more curious about why teachers believe what they believe, including learning styles. Math teacher Dylan Kane provides a great example of this curiosity in action. He recently conducted a short, non-scientific poll of his followers on Twitter — many of them educators – to learn more about where enthusiasm for learning styles might stem from.

The contrast suggests that while learning styles are popular, what’s really popular is instruction involving multiple modalities. Perhaps some teachers who express a belief in learning styles “really mean that they try to use a variety of representations and activities in class,” as Kane wrote in a subsequent blog post. This is something to encourage in education; good teachers know the value of teaching their students in more than one way.

Of course, Kane’s poll was not scientific, and we don’t know how many of the respondents are practicing teachers. But we suspect most advocates of learning science advance their arguments against learning styles with even less data regarding the existing beliefs of their intended audience (educators). If so, their attempts to build new knowledge in educators may be premised on a misunderstanding of what educators believe, or why. That’s a recipe for an unproductive dialogue.

2. What scientific insights about learning are important for educators to understand?

Another principle from cognitive science is that our decisions are guided by mental models and representations. The most common forms of science communication focus on making evidence-based information available, and assume this information will be incorporated into the recipient’s mental model as a matter of course. Usually, it isn’t, and recipients simply retain their existing ways of seeing the world.

Instead of simply sharing evidence or information, we suggest advocates of learning science spend more time helping teachers understand models of learning they can employ in the classroom. Happily, many scientific principles are useful for teaching, but here we consider just one: dual coding.

Dual-coding theory states that the mind processes words and pictures along different pathways. Researchers have found that instructors can present more information to students by distributing it across both words and pictures. Instruction that does this in artful and in complementary fashion often will reach more students than instruction that does not.

As some learning scientists have aptly observed, dual coding covers similar territory as learning styles. For that reason, advocates of learning science would be wise to introduce dual coding as an alternative to learning styles. We doubt teachers will immediately reject learning styles – students rarely discard their initial (mis)conceptions right away – but over time, if dual coding proves effective in the classroom, teachers may find that learning styles has lost its appeal. Once again: it is better to replace ideas than to debunk them.

3. How might we create opportunities for teachers to practice their understanding of learning science?

Practice is essential for learning, but not all practice is equally effective. A great science teacher (or teacher of any subject, for that matter) provides students with many opportunities to practice their new understanding in structured ways.

This presents a real challenge for advocates of learning science. Blog posts, op-eds, and social media all have a role to play in raising scientific awareness, but we know that real learning requires more. How can we foster opportunities for teachers to try new science-informed practices – and receive useful feedback as they do?

There are no easy answers to this question, but our hunch is that at a minimum it will require learning scientists to approach educators with more humility. Instead of attacking myths, scientists need to approach educators as professional colleagues. As colleagues, teachers and scientists have much to learn from and teach each other about what works in a classroom.

Learning scientists can help practicing teachers improve by inviting them to attend learning-science conferences, collaborating on rapid-cycle research projects, and by providing direct professional development in the local schools where teachers teach. And whenever these interactions take place, we hope learning scientists will listen to teachers, and learn from their experiences in the classroom to inform future research.

This sort of approach might reach Terry J. in a way that info-spreading never could. Terry loves science, research and education. It’s hard to imagine her saying “no” to an opportunity to collaborate with scientific researchers in a collegial way. Will anyone offer her the opportunity?

Building bridges between science and teaching

We believe we are at a crucial moment in the relationship between learning science and education. More than ever, there are organizations and individuals seeking to share the fruits of scientific research with educators. We suspect there have never been as many books about learning science in the hands of teachers as there are today. We should celebrate this development.

At the same time, we worry about the danger of backlash. “Teachers must reject the learning styles ‘neuromyth’” is a provocative headline. It’s also polarizing. The more teachers are told what they must do or believe, the greater the risk that they will become antagonistic toward learning science, or even research generally. We should endeavor to prevent that from happening.

Scientists know a lot about the cognitive processes that can lead to learning – or not. Educators know a lot about the instructional processes that can lead to learning – or not. By building on their respective knowledge bases, and treating each other with mutual respect, we can foster the scientific profession of teaching.

4 thoughts on “Why Mythbusting Fails: A Guide to Influencing Education With Science

  1. I fully endorse the principle that people who want to engage teachers should embrace humility and avoid condescension, and actually I would like to press that point a bit. Although it urges cognitive scientists to change their communication techniques, this post seems to accept their premise that
    teachers are clinging to wrong science.

    Nonetheless you yourself point out that teachers’ stake in the “learning styles” concept is not necessarily related to what the cognitive scientists are criticizing. To your point about dual coding, one of the big things teachers have at stake in the notion of learning styles is that students benefit from a diversity of information processing pathways. The other big thing imo is the basic insight that different students respond to the same instructional techniques differently. Neither of these is a controversial point and I’m sure the cognitive scientists would instantly grant both. Their criticism seems to be aimed at three more involved premises:

    (1) That all or most individual learners have a single dominant learning modality,
    (2) That it is fixed over time, and
    (3) That it can be accessed by self-report.

    I mean what do I really know, but honestly I doubt that there are very many teachers to whom any of these three things are very important.


    1. Although it urges cognitive scientists to change their communication techniques, this post seems to accept their premise that teachers are clinging to wrong science.

      So what exactly is happening here? Because, as you note, it’s not hard to see why a theory that entails multiple-modalities, individual accommodations and the use of visuals is going to be attractive to teachers.

      Let’s talk for a second about Cognitively Guided Instruction. CGI is now a theory of teaching through problem solving, but it didn’t start out that way. When Carpenter and Friends started out, it was an investigation into how students approached different addition and subtraction word problems. In fact, they developed an entire theory as to what contextual factors trigger different sorts of strategies from students. (More on this here.)

      So, they developed this theory about how kids solve word problems. Now what?

      The awesome thing about the CGI researchers is they moved to the teachers. Teachers, after all, spend a lot of time watching kids solve word problems. Surely they must have an understanding of some of these things? So the CGI team moved from studying how students think to studying hos teachers do.

      As I understand it, the researchers discovered that teachers did have a lot of correct thoughts about how students think about word problems. Their learning, though, wasn’t systematic. So maybe teachers knew that some problems would be easier or harder for kids, but didn’t quite know how to explain why. Or that kids would tend to use counting for this story problem but not that one; likewise, teachers didn’t have an organized explanation for when, how and why kids tackled word problems.

      Ever since I learned this, this has been my go-to way of understanding the relationship between teacher knowledge and research knowledge.

      What I think is happening with learning styles is that teachers have all this informal knowledge about visuals, individual differences between kids, the need for multiple approaches to teaching the same idea. These ideas, typically, are not especially well organized. Teachers know that visuals are sometimes important, but it’s just sort of an instinct. We don’t have a systematic way to think about how visuals are important, why, when to use them, etc.

      And then, one day, a new idea comes along. Learning styles ties all these things together, it has a research pedigree, and it starts answering some of the whys, hows, and whens.

      All of this is to say that learning styles is like an organizing capstone that’s built on a TREMENDOUS amount of informal, disconnected teacher knowledge. All the informal knowledge is correct. The issue is that the capstone isn’t. The ways in which the learning styles hypothesis organizes the informal knowledge has, according to our best science, turned out to be false.

      The challenge, then, is building a better organizational theory that can help teachers make sense of all of this correct, informal knowledge about learning in a classroom.

      Now, the thing researchers often don’t get is that learning styles is just the tip of the volcano. (Need to start phasing out those iceberg idioms, it’s seeming.) And so they don’t engage with it. This leads to teachers feeling disrespected, because the learning styles hypothesis comes to represent ALL of this other knowledge that it helps organize. And so — in addition to the autonomy issues — scientists can sound like they’re contesting teacher knowledge itself. And that makes teachers angry and defensive, and that leads to learning styles getting actually believed over and above the uncontroversial parts of teacher knowledge.

      That’s why, in the piece, I’m trying to encourage people to BUILD on the existing teacher knowledge. Offer something like what the CGI people offered — a way to make sense of classroom reality. A better scientific capstone.


  2. Oop, two addenda:

    I. A minor correction: to the sentence “…is the basic insight that different students respond to the same instructional techniques differently…” I should have also added “… and the same student responds to different instructional techniques differently.”

    II. I thought of a third thing that I think teachers have at stake in the “learning styles” concept, and this one might be more controversial with some of the cognitive scientists, but it is very unclear to me in this third case that the cognitive scientists have any privileged access to information on the matter. Namely: I think many teachers see the notion of learning styles as a way to acknowledge the validity of the diversity of ways students engage with the classroom. E.g. to name the who can’t sit through a chalk and talk lesson but loves to come up and act out addition with a little dance as an equally valid member of the learning community as the kid who can sit still and track a speaker for 15 min straight.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s