Thursday, February 25, 2010

Learning Styles: What's Being Debunked

I have a response to an article on learning styles from Teacher Magazine
Here is the original article, and here is the link for my response, which you need a username and password to the site for, so the gracious editors have allowed me to post it here as well. By the way, you can sign up for a free account, which I would recommend if you are interested in this sort of stuff. So, here is my piece:

In her recent article “The Bunk of Debunking Learning Styles” Heather Wolpert-Gawron makes a plea for common sense in the face of research findings that contradict her direct observations of learning styles in the classroom. She cites a recent article (“Learning Styles: Concepts and Evidence” by Harold Pashler, Mark McDaniel, Doug Rohrer and Robert Bjork, in the journal Psychological Science in the Public Interest) which has claimed that there is no scientific evidence that learning styles exist, and argues that knowledge she’s gained during her 11-year career in the classroom prove that they do. As a psychological scientist and a son and husband to classroom teachers, I feel the need to respond.

First, it is necessary to clarify the definition of learning styles and the predictions of learning styles theory. Second, I want to pinpoint what the “debunkers” in question are claiming, which I think is more specific than Ms. Gawron-Wolpert describes. Finally, since she seems to believe that basic science is useless when it comes to the practice of teaching, I want to describe how basic cognitive science can apply to teaching. Although the scientific search for evidence of learning styles has yielded no evidence of their existence, basic psychological science can help teachers, even as it steers clear of dictating exactly what works in any individual classroom.

Learning Styles Defined

We must begin with how learning styles have been defined, both in the research literature as well as in educational practice. Learning styles theory does not propose generic differences between how students learn, but asserts a specific kind of difference. A learning style, by the prevailing account, is a preferred mode of learning, distinct from ability, and independent of content area. For example, a visual learner is not necessarily better at learning math or geography than other students, but a better learner when any material is presented visually, compared to other modes of presentation. This may not be Ms. Wolpert-Gawron’s definition of learning style, but it is the definition used by researchers for over 50 years, as well as the educational policy makers who are currently implementing learning styles theory. For example, although multiple intelligences may seem similar to learning styles, Howard Gardiner has made it quite clear that multiple intelligences is a theory of abilities, not of styles. The current learning styles theory defines “mode of learning” as a preferred sensory channel, either visual, auditory, or kinesthetic, but there have been many ways of defining “mode” in the past.
Why is it critical that a learning style be distinct from content and ability? Because one important claim of learning styles theory is that no one learning style is superior to another. If visual learners learned math faster, and kinesthetic learners learned basketball faster, we wouldn’t need to label them with learning styles at all, we could say that one group has mathematical aptitude and the other athletic aptitude. Unlike decisions about what works in any given classroom, which are for individual teachers to make, learning styles is a theory of how the mind works, and it is framed in a way that makes it suited to controlled scientific testing. The key scientific claim for learning styles theory is that we could teach two classrooms of randomly assigned students the same content, but one would be taught “visually”, and one “auditorially.” The visual learners should do better than the auditory learners in the visual classroom and vice versa in the auditory classroom. If everyone does better in the visual classroom, then we would conclude that the content is more suitable for visual presentation. If the “visual learners” do better in both classrooms, then you have identified an ability, not a style. This “matching styles to instruction” pattern of relative differences in learning is the evidence that the authors of the paper above searched for in the scientific literature. Several studies claim to support learning styles, but did not perform this critical test. Those few that did satisfy this design failed to find evidence for learning styles.

What the Debunkers Do, and Why

These researchers have identified the central claim of learning styles theory, and failed to find any scientific evidence for this particular claim, despite many relevant studies. Ms. Wolpert-Gawron accuses them of invalidating the practice of differentiating learners at all. She suggests that they don’t mention the “alternative —that of teaching all students the same way.” This is not the alternative that the scientists have in mind. One representative quote from the article is, “it is undeniable that the instruction that is optimal for a given student will often need to be guided by the aptitude, prior knowledge, and cultural assumptions that student brings to a learning task.” In other words, obviously students differ, just not by learning style.
To refute her view of the debunkers, Ms. Wolpert offers many examples from her own experience which show that learners are different. While this is not the claim of the scientists, why should they bother debunking this theory at all, if many people define it as generally as Ms. Wolpert? Those scientists who debunk learning styles do so in order to remove the obstacles to teachers’ focusing their attention on dimensions of learners that both science and practice have identified as critical. In their words, “assuming that people are enormously heterogeneous in their instructional needs may draw attention away from the body of basic and applied research on learning that provides a foundation of principles and practices that can upgrade everybody’s learning.” Learning styles theory distracts teachers from principles and practices that we all agree are successful.
Ms. Wolpert-Gawron clearly agrees, stating that the engagement of all students is crucial to learning, but she maintains that a learning styles approach fosters attention to student engagement. Perhaps this is true for the way that she has defined learning styles, but it is not true for learning styles theory as a scientific theory of mind, as it is applied to many teacher evaluations, or state standards. Enforcing attention to learning styles directs teachers to a particular method of student engagement, and necessarily away from another. For example, to illustrate a certain concept, one could tell a very engaging story, related to the lives of the students themselves. But if one were a teacher with intense time pressure, meetings galore, and multiple classes to prepare (which is to say, any teacher), learning styles theory would encourage attention to the sensory modality of the story (visual, auditory, or kinesthetic), rather than to the meaning of its content, its intrinsic interest and its appropriateness for the particular lesson of the day. This doesn’t seem to trouble many teachers, who like Ms Wolpert-Gawron have been happily ignoring the central claims of learning styles theorists. However, this may not be the case with beginning teachers, or teachers who are stringently evaluated by arbitrary criteria based on the myth of learning styles.

The Role of Basic Cognitive Science in the Classroom

The misunderstanding of the scientific claims does undermine the subsequent article, but as a cognitive scientist who often reports research findings to my family of teachers, I feel it is important to address and confront the gaps that she mentions (and doesn’t mention) between research and practice, as well as her clear disdain for the scientists in question and their unwelcome incursion into her classroom. She is not unique in this attitude, nor is it limited to learning styles. This gap between basic research in cognitive psychology and the practice of teaching has negative consequences for each side. In their distrust of basic science, teachers miss an opportunity to improve their students’ learning by applying their expertise on relevant dimensions of learning. In allowing this distrust to exist, scientists undermine the public’s trust in the value of the basic science to understanding human behavior. Just as the science of medicine need not undermine the expertise of a doctor, the science of psychology need not invalidate practice-based knowledge, but rather supplement it with general information about theories of the mind and learning, without direct prescriptions for what to do in a certain classroom situation.
In order to repair this distrust, scientists must first summarize our findings for audiences outside of our community, with an eye toward informing educational practice. In doing so, we need to describe our basic science findings as theories of how the mind works, not straightforward recipes for educational reform. Daniel Willingham’s recent book “Why Don’t Students Like School,” may not do a great job answering the question in the title, but it serves as an excellent summary of consensus views in cognitive science as they apply to education (the learning styles and multiple intelligences chapter is particularly cogent and insightful). But we can’t stop there. We must also dispel myths, and we in psychology have a larger set of myths to dispel than others. When these myths exist, they are corrosive to science, because while seeming to represent science (“well, it says it’s a theory”) they do not provide the measurable, reliable results that science demands. These myths are perpetuating identity theft of science, calling themselves science and wrecking havoc on our credit scores, yet many scientists don’t connect the bankruptcy of public trust in science with the myths that we let roam freely. In the case of learning styles, minimal evidence has been exaggerated and marketed to educators and administrators, outside of the checks of the scientific process. As scientists we must take greater efforts to reign in this misapplication of science. The recent article on learning styles that Ms. Wolpert-Gawron refers to is an example of this, as is another excellent book “50 Great Myths of Popular Psychology,” by Scott Lilienfeld, Steven Jay Lynn, John Ruscio and Barry Beyestein (learning styles is number 18). What’s more, the journal in which the learning styles article appeared, “Psychological Science in the Public Interest,” is a journal with the laudable goal of improving “giving psychology away,” with both topics and authors commissioned after a careful nomination process.
In addition to summarizing the scientific consensus, and dispelling myths, the basic science of learning should clearly state the questions that we do not know the answer to, and get out of the way of expert teachers. Experienced teachers certainly have knowledge that science does not. Given that practice-based knowledge is practical knowledge, gained by classroom experience, it can sometimes be specific to the population a teacher serves, rather than a general knowledge of how people learn (just like baseball players are not necessarily experts in the general rules of projectile motion). It is not basic scientists but political reformers who are turning scientific theories into coarse criteria for evaluating teachers based on test scores, or a simple checklist. Basic psychological scientists are in general cautious, as well as skeptical of attempts to directly apply general theories to particular classroom situations. What Ms. Wolpert-Gawron is interpreting as science telling her what she sees in her classroom is in fact a summary statement of scientists telling her what they don’t see, despite having looked in the best ways they know how. The authors of the study, in my mind, are attempting to empower teachers to use the principles of learning that they know work, while encouraging them to steer clear of myths, which may have had a scientific-seeming provenance (if it’s from Harvard…), but have not received rigorous scientific support for critical claims.
I argue that basic science can concern itself with general mechanisms, and teachers can practice applied science in their own classrooms, but what happens when there seems to be a direct confrontation? How do we decide between the scientist in his sterile lab vs. the expert teacher with 11 years experience and 2500 students? In other words, why should experienced teachers let scientists tell them what is and isn’t a myth when common sense dictates otherwise? Because despite the fact that personal experience is very compelling and convincing, human beings are notoriously bad at direct observation of complex relationships. Our stone-aged brains notice patterns that aren’t there, seek out evidence that confirms our preconceived notions (called the confirmation bias), and ignore evidence that might prove us wrong. This is just as true for surgeons and scientists as it is for teachers, and the controlled observation, whether in a scientific lab, or through a double-blind study, or using randomized assignment to experimental groups, is absolutely critical element to the success of science in explaining and predicting complex human phenomena. This holds equally true whether it be the spread of disease or the process of learning. The history of common sense has been remarkably wrong, even in those experts who have seen thousands of cases. Scientists are people too, and so we don’t trust our own observations any more than anyone else’s, but rather use them to inform what should be tested in a controlled study. If controlled study after controlled study fails to observe, or offers contrary evidence to our most cherished beliefs, we have no choice but to give them up.
The goal then, is a collaboration to arrive at the most relevant dimensions in learning, and the most effective way of teaching, respecting the expertise of the teacher, but accepting that in some cases, science can point out where myths exist. For example, the science of cognitive psychology can point to the necessity of practice (for example, through drilling) and background knowledge for deeper learning as well as the ways in which motivation and engagement are critical to learning, but cannot offer an ideal way to balance these two in a American History lesson for English Language Learners. The science can note that there is considerable evidence for the organization and meaning of a lesson having a large effect on learning, and little evidence that the color of the ink, or whether the words are on a page or on a blackboard have any effect. This does not mean that it doesn’t matter in any classroom, just that teachers should be cautious in choosing to spend time on choosing the color of the ink and err towards thinking about the structure and meaning of their lessons. However, only the teacher can apply these considerations to their subject and the particular students in front of them. As Ms. Wolpert-Gawron notes, teaching learning styles is far more difficult than not, but science here is trying to offer a way to make things easier. Just as science in medicine can call a doctor’s attention to a set of manageable indexes of health, science in education should aim to suggest to our teachers a set of relevant dimensions of learning, with the understanding that teaching is immensely complex. While basic science can offer theories and insight into how learning works, no one knows the particular students in front of her better than the experienced teacher.


Sanjay Srivastava said...

Great article! Seeing how slowly evidence-based practice has made inroads into psychotherapy -- an applied domain that has historically had much closer ties to scientific psychology -- I'd say that advocates for evidence-based education have a tough task ahead of themselves. But certainly an important one to pursue.

Unknown said...

Boy, Cedar, this is a great article ... but it reminds me of teaching Child Dev ("my parents did that and I turned out just fine"). I'm glad you wrote this and I'll forward it on to some non-psych folks, but wow. Convincing people of science over experience is tough. In fact, Sanjay's comment reminds me of how hard it is sometimes to have scientists jump over this.

Sam said...

Great piece, Cedar. Really nuanced argument for attention to detail and data in education.

Liz Ditz said...

Hello, Cedar -- John Wills Lloyd sent me over.

The educational community is sometimes depressing in their (collective) response to research and science. Equally depressing is the rather low standard of evidence in educational research, especially at the k-12 level.

In the comments at Teachers' Journal, Heather Wolpert-Gawron wrote:

By the way, what are your thoughts on Judy Willis and her studies of neuroscience in regards to education?

This is a problematic question, as to the best of my knowledge Willis hasn't published anything in the peer-reviewed scientific literature on neuroscience and education.

It turns out that Judy Willis is a former neurologist who retired and became a teacher. She has published several books and now a tours on the professional development (PD) circuit. If I remember correctly, she retired from neurology in 2004 or thereabouts and has been teaching since about 2006 (but I could be wrong about the dates by a few years or more).

What Willis does do is to use her unusual position of being highly educated in neurology to publish translations (so to speak) of research in cognitive science into forms classroom teachers (and parents) can use.

Willis seems more balanced and restrained that some of the other "brained-based learning edufolk, but I'm still, well, skeptical. One of the things that makes me skeptical is her approach to teaching reading in kindergarten & first grade, as expressed in this article. In other words, I suspect she has an ideological axe to grind when it comes to pedagogy.

I would, however, recommend to you the International Mind, Brain & Education Society web site

Another enterprise is the Learning & The Brain conferences, twice yearly, once on the East coast and once on the West coast.

Cedar said...

Thanks Liz, for commenting. Looking at Judy Willis' website, I was a little skeptical, but it seemed more reasonable than some of the "brain-based" education stuff.
I looked over that article you posted of hers, and although it is critical of phonics instruction as a good thing regardless of the context, I thought it sounded measured and reasonable. I personally don't think you need to know when the amygdala lights up to know that it is misguided to favor
"an approach that puts phonics first at the expense of intrinsic appeal and significance to the young reader."
Making things interesting is so important, and you don't need any fMRI study to tell you that.
I am a fan of IMBES on facebook, and I correspond regularly with Dan Willingham (who was one of my graduate advisors).
That conference looks good, hopefully I'll make it one of these years.

Michael 4MAT said...

Nice response to this research article. I wanted to add the specific referenced from the researchers on their finding.

I believe that many are misinterpreting this research that is the source of this. Here is the cite from the researchers on their review of literature from the field.

"Our review of the literature disclosed ample evidence that children and adults will, if asked, express preferences about how they prefer information to be presented to them. There is also plentiful evidence arguing that people differ in the degree to which they have some fairly specific aptitudes for different kinds of thinking and for processing different types of information.

However, we found virtually no evidence for the interaction pattern mentioned above, which was judged to be a precondition for validating the educational applications of learning styles."

In their research, the researchers were attempting to verify, that if:

a. you determined student style (visual, auditory, kinesthetic),
b. employed a strategy to appeal to that style,

Did this have an impact on standardized test scores.

Those in the field would call this "matching."

Many others in the educational field argue for broadening of instruction to address student learning differences, or multiple forms of instruction.

This research was not looking to evaluate that kind of approach (broadening). So statements like learning styles debunked is most definitely an over-statement of the research finding.

Cedar said...

Thanks for commenting.
I agree that the researchers specified a relatively narrow claim, and summarized the research that addressed that claim. But the reason that they did this is that this claim is the linchpin of the learning styles argument.
I agree with you that broadening the curriculum is a good idea, and presenting content in a variety of modes is also a good idea. But I do not think it is a good idea because you will get the visual learners with the video, and the audio learners with the lecture, and the kinesthetic learners with the interpretive dance. "Broadening" as you call it, helps maintain everyone's attention by changing things up.
The authors don't address the merit of broadening on the basis of learning styles, because they identify the meshing (or matching) results as a precondition for proceeding using learning styles. Students do differ (in aptitude and background knowledge, among other things) but if you are going to design curricula and teacher evaluation tools (or simply advocate broadening) using a certain dimension of student learning difference, I would advocate confirming in the lab the existence of that difference. If not, by all means, go ahead and broaden the curriculum (I am an adamant defender of a broad curriculum in higher ed, as you could probably tell by some of my recent posts) but don't say that it is based on cognitive science.

Catherine Scott said...

Re: teaching reading, who set up the utterly false dichotomy 'intrinsic appeal' versus 'teaching phonics'? Why do we assume that these techniques are a put off for kids?

We seem to have accepted the association between phonics instruction and task master style teachers flogging kids with canes as they drill them on meaningless facts until their eyes roll back.

Kids positively value increasing their competence, by learning how to decipher text, for instance. The right sort of phonics (synthetic phonics) properly taught does that for them.

Cedar said...

@gardener cath: good point. Part of my purpose in debunking learning styles is letting teachers be interesting on their own, without giving them overly prescriptive recipes. I did not mean to degrade phonics, or teaching a rich curriculum of facts, which I believe can be quite interesting. I agree that these things often get an unjustified bad reputation.