No, I'm not having second thoughts about about school relevant EAL which I remain just as committed to as ever. I'm talking, rather, of the 'learning styles' (LS) fad and how it continues to lead us perilously astray. For those unfamiliar with this whole learning styles business I'll begin with a few details to put the topic into some kind of historical context and clarify terms. I'll then look at the 'supposed' evidence that supports building lessons around students' learning style preferences. For a detailed overview of the issues involved I recommend the comprehensive Understanding Learning Styles: Making a Difference for Diverse Learners (2010) by Allen, Scheve and Nieter.
To begin with a timeline: According to Coffield et al (2004), academic interest in learning styles really only emerged in the 1950s. An internet search identifies four studies addressing the topic during this decade. Seven studies appeared in the academic literature during the 1960s, and then 21 studies in the 1970s. Only in the 1980s did Learning Styles really begin to grab our attention. By this time the term 'learning style' had settled on its current definition of whatever a student identifies as his or her preferential means of learning, given a choice. For the last 30 years or so teacher training courses have covered 'learning styles' in some depth, promoting the notion as key to delivering lessons that truly differentiate between different student's needs and which result in more learning per 'unit' of instructional time.
But why the the uptick in interest following the lackluster 1950 and 1960s? After all, who has ever doubted that students might prefer to learn in one way or another? In fact the the whole LS topic might never have attracted much attention at all were it not for the claim (formalized as the Meshing Hypothesis) that students will learn more from studying in the style that best 'suited' them than from a style they disliked or were ambivalent to. This notion piqued teacher interest touching, as it did, upon our instinct to impart knowledge in as timely a manner as we can manage. The hypothesis drew its rationale from Gardner's influential Frames of Mind: The theory of multiple intelligences (1983). This text had already proved a hit with teachers and academics and found a place on staffroom bookshelves and reading lists for trainee teachers. Despite that Minds focused on intelligence (not learning), by the mid 1980's learning styles and MIs had for most educators become opposite sides of the same coin -or, as the Center for Innovative Teaching and Learning put it, become "conflated." Gardner bitterly regretted the conflation and spoke out against this entanglement at every opportunity. Despite his protestations, however, even the briefest perusal of the '80s and '90s literature reveals LSs and MIs becoming ever more conjoined, the one notion nurturing and gaining sustenance from the other. As MI attracted converts, so teachers began to acknowledge learning styles in their day-to-day teaching. Barbara Prashnig of the Creative Learning Company (Aukland) put it like this:
When talking with professional educators about new teaching methods I often hear comments like “In our school we have introduced Multiple Intelligences which now cater for our students’ learning styles” and “When I teach in many different ways, using MI, I am covering all my students’ learning needs” (2005).
So, what exactly did Minds have to say of on learning styles? Surprisingly little, in fact. Frames of Mind explored notions of human intelligence. How we actually learn via that intelligence scarcely receives a mention. Gardner's aim was simple: to challenge the notion of a single 'general-purpose' problem solving capacity -the all embracing g-factor that IQ tests purport to measure. Unlike most of his his contemporaries, Gardner argued that we have not one but several forms of intelligence meaning that, for example, a learner might demonstrate impressive ability (high intelligence) in one field of endeavor -perhaps a child displayed a astonishing mastery of mathematics- and yet less so in another; the same child might have little aptitude, say, for playing musical instruments. This was wonderfully egalitarian, to say nothing of intuitive. In no time, it seems, the hypothesis had found favor among the teaching community and begun to make its way into teacher education courses and training manuals. Faith in the traditional IQ test for categorizing children according to academic potential now seemed misplaced at best, and perhaps even seriously flawed. If Gardner had got it right, the argument went, then a child might score poorly on the 11+ (to name but one popular IQ measure) yet display very real potential in other fields that the test didn't happen to assess. If nothing else, Gardner had at least issued a timely reminder that giftedness can express itself in forms other than aptitude in the verbal/logical reasoning that schools have long valued so highly.
So, what then, are the intelligences that Gardner claimed to have identified? Well, he claimed 8 in all, but didn't rule out the possibility of others. In fact, he felt he suggested a ninth: existential intelligence but stopped short of listing it with the others. The diagram below sums up the main types:
Frames (1983) proved both controversial and intriguing. In the heated debates that followed, contemporaries to pointed out a lack of school based evidence along with little by way of experimental support from cognitive neuroscience. Some 20 years after publication you might have thought the facts would have emerged and loose ends resolved. Alas, No! The topic remained as controversial as ever -so much so that, writing in 2004, Gardner conceded that "MI theory has few enthusiasts among psychometricians or others of a traditional psychological background" despite what he felt was an impressive body of supportive evidence. The difficulty Gardner faced, as Visser et al (2006) explain, lay largely in the correlations of each of the discrete intelligences with 'g' (i.e. general intelligence), together with the (often) strong correlations between the intelligences themselves. Faced with such correlations psychologists saw no good reason to suppose the intelligences were autonomous of one another.
Our analyses of tests measuring the “intelligences” of Gardner's MI theory revealed that many of those tests were substantially intercorrelated, despite representing different domains of Gardner's framework, and also showed strong loadings on a g-factor and strong
correlations with an external test of general intelligence. These results are difficult to reconcile with the core aspects of MI theory (Visser et al, 2006).
The 'g' factor might have taken something of a hit during the 1990s but Gardner had failed to deliver the knockout blow he'd intended. By the early 2000s, as now, the reputation of IQ tests as measures of problem solving ability remained firmly intact. The tests continue to predict, and with some success, how children would perform in national exams.
Notwithstanding the above, no one denied that MIs had obvious potential to explain the phenomenon of 'giftedness' -i.e. a particular aptitude for a particular skill. We can probably all recall the child from our schooldays whose artwork proudly adorned the hallway and yet who languished in the bottom set in Mathematics. If we assume, with Gardner, that this particular student possessed high artistic intelligence and relatively low intelligence for mathematics then this makes perfect sense. Visser et al (2006), and other MI critics did not deny that some of us display astonishing skills -artistic, musical or otherwise. They argued, rather, that such giftedness does not challenge the impressive power of 'g' to predict success in school generally (at least as measured by formal exams), or our general problem solving activity. Gardner had little to say on the practical issue of how to measure giftedness but, then again, perhaps this didn't matter in a practical sense. After all, giftedness tends to announce itself clearly in lessons, on the sports field, or in the playground. Devising tests to quantify different intelligences still remains a controversial and not altogether successful effort.
The evidential shortfall for MIs and its implications seems lost on teachers and school management. By the early 1990s Fleming and Mills (1992) had taken the MI theory one giant step further, pushing the case for discrete learning styles in every classroom. The idea seemed sound enough: just as children might have different aptitudes, or intelligences, so they might learn more effectively if teaching matched their preferential learning method (or, teachers estimation of what that might be). Soon after, and to Gardner's dismay, researchers had come up with any number of assessment instruments claiming to determine a child's preferred learning manner, what category of learner the child fell into and how children would approach learning tasks given their unique cognitive makeup. The well known VARK instrument, for example, identified 4 such styles:
1. Visual learners:
To rephrase Callie Malvic, these make up that set of students who learn through sight. Put another way, they "understand information better" when when teachers prevent it in a "visual way." She adds: "These are your doodling students, your list makers and your students who take notes."
2. Auditory
Auditory learners tend to learn better when the subject matter is reinforced by sound. These students would much rather listen to a lecture than read written notes, and they often use their own voices to reinforce new concepts and ideas.
3. Reading/Writing
Reading/writing learners prefer to learn through written words. While there is some overlap with visual learning, these types of learners are drawn to expression through writing, reading articles or books, writing in diaries, looking up words in the dictionary and searching the internet for just about everything.
2. Kinesthetic learners
Kinesthetic learners learn through experiencing or doing things. They like to get involved by acting out events or using their hands to touch and handle in order to understand concepts.
As with MI, the evidence for learning styles has proven elusive but -arguably rather more so. By the early 2000s some openly called learning styles the new neuromyth (see e.g. Centre for Educational Research and Innovation, 2002). Children might prefer to learn in one way or another, true enough. But as Derek Bruff (2011) stated at the Lilly Conference on College Teaching, the evidence suggested teaching to a preference does not appear to positively impact upon leaning outcomes. This wouldn't have struck the audience as a surprise. As Pashler, McDaniel and Rohrer had pointed out some two years earlier, those studies claiming benefits from acknowledging students' learning styles all too often embodied serious methodological flaws.
If you're wondering what Gardener had to say on learning styles, clearly he viewed the notion as a step too far. In 2013, he confided that the topic had driven him "to distraction." He critiqued the learning styles notion on two grounds: First, the whole notion failed the test of consistency:
'Those who use this term do not define the criteria for a style, nor where styles come
from, how they are recognized/assessed/exploited.
Second, and turning to results:
When researchers have tried to identify learning styles, teach consistently with those s styles, and examine outcomes, there is not persuasive evidence that the learning style analysis produces more effective outcomes than a “one size fits all approach.”
And its not just among child learners that teaching to a learning style has proved unproductive. Disappointing results have also emerged from studies involving graduates and adult learners (Newton, 2015). Of course, an absence of evidence does not disprove a hypothesis as such -for that we need counter-evidence. So what of the counter-evidence, then? Perhaps the last word on the current(ish) state of play should go to Joshua Cuevas (2015) who examined the literature since 2009 after weeding out studies that did not conform to the very particular guidelines that Pashler et al (2009) argued future researchers needed to acknowledge if they hoped to verify the learning styles hypothesis. Cuevas concluded that while not all studies wrote off the hypothesis as 'unhelpful' the more methodologically sound studies found little evidence in support. Indeed, the majority of research evidence suggested that efforts to match learning styles to teaching "had no benefit to student learning."
Why so popular?
Given the paucity of evidence, it seems odd that so many teachers still buy into what Kirschner (2016) calls the learning styles myth. But buy into they have. Today we have a cottage industry busily churning out texts, assessment instruments and bespoke resources targeted to students of one supposed learning style or another. And that's not to mention the online courses and conferences on the topic held in the name of progressive teacher training. According to Dekker et al (2012), no less than 80 percent of their sample of teachers from the Netherlands and UK believed that students learnt better when teaching acknowledged a learner's preferred learning style. The Wellcome Trust (2013) found a substantial "76 percent of teachers claimed to use" learning styles in their daily teaching practice. Headteachers, seem particularly enamored (gullible?). Quoting correspondence with professor Bruce Hood (of Bristol University, UK), The Educator (March, 2017) referred to "a recent poll of more than 100 head teachers" that found over 85% believed that acknowledging learning styles had educational benefits. While the poll limited itself to British independent schools, Newton and Salvi (2020) caution that the learning styles myth has now gone global. The authors examined 37 studies representing 15,405 educators from 18 countries around the world, spanning 2009 to early 2020. The study concluded "There was no evidence that this belief has declined in recent years." Moreover, "95.4% of trainee (pre-service) teachers agreed that matching instruction to Learning Styles proves effective."
So, why the overwhelming popularity when the evidence seems so lacking? the Teaching Revolution website believes the answer lies partly in intuition: namely, the idea that everyone has a distinct preference for learning has an intuitive appeal. We all differ in temperament, mental faculties, physical prowess, after all, and so it makes sense, surely, "to take those differences into account." Learning Style adherents would claim they do just that. The article goes on to suggest that the learning styles notion also lends itself to simple assessments of a sort that students find fun - "a huge draw," the authors note, "for students and teachers alike." Equating good teaching with fun has a long history. Only now, it seems, have a few brave souls stepped forward (finally!) to question whether 'fun' really delivers. The third argument comes from Kaufman (2018). He sees teacher support for learning styles stemming from the most laudable of motives -"care for students." Unfortunately, Kauffman argues, once we add to this the distaste teachers feel for "immutable' characteristics" you end up with all the ingredients from which something as warm, "fuzzy," but damaging, as learning styles can flourish. Kaufman limited his remarks to American schools, but his message will hit home with many of us in international schooling and those in the UK.
Conclusion audio-lingualism
If teachers indeed jumped on the bandwagon to embrace the learning styles 'neuro myth' ( CERI, 2002) with reckless haste, it's hardly the first time we have embraced a practice only to be let down by later research. Take our own field of EAL, for example. In the 1950s audio-lingualism was all the rage and continued as a popular teaching method right up until 1970s. And yet, as Chomsky showed in Syntactic Structures (1957), the whole audio-lingual edifice presumed features of human language, and acquisition, that were patently incorrect. As with audio-lingualism, teachers today seem remarkably tenacious in clinging to notions that psychologists view as seriously flawed. The lag between discovering a practice is outright wrong, and teachers giving it up, can seem distressingly large. The rise and fall of audiolingualism is a case in point.
For SR-EALists, the topics of Learning Styles cannot but attract our interest and it deserves serious consideration. The essence of SR-EAL lies in teaching approach that meets the particular needs of students in particular schools. That children might have different learning styles is something that any self-respecting SR-EAList naturally wishes to look into. While no one doubts differences in student abilities, at the present time attempting to teach according to learning style preference seems misguided and somewhat wasteful of time and resources. Perhaps this will change. The teaching profession today still sees merit in the learning styles notion, so let's keep an open mind. Could it be that somewhere lies a basic truth that the psychologists have yet to discover or overlooked. Perhaps. As it stands, however, the jury has delivered its verdict and, until we have a retrial, that verdict comes across as unambiguous and somewhat damning: “As the critics of learning styles correctly claim, the meshing hypothesis (matching instruction to students' learning styles maximizes learning) has no rigorous research support" (Felder, 2020).
References
Felder, R. (2020). “OPINION: uses, misuses, and validity of learning styles,” in Advances in Engineering Education. Available online at: https://advances.asee.org/opinion-uses-misuses-and-validity-of-learning-styles/ (accessed September 01, 2020).
Philip M. Newton and Atharva Salvi. "How Common Is Belief in the Learning Styles Neuromyth, and Does It Matter? A Pragmatic Systematic Review." Frontiers in Education (First published: December 14, 2020) DOI: 10.3389/feduc.2020.602451
Comments