Putting the Elephant Back Together Again

William Coleman

November 17, 2020

How Learning Styles Meet Their Perfect Match With Machine Learning and Digital Education

            Everyone wants the best educational experience for their child, but with so much uniformity and monotony in the education system, it can be really hard for some to fit in to the traditional classroom setting and get the best experience possible. Although some schools may boast of turning textbooks into tablets and lengthening daytime breaks, nothing has fundamentally changed for those who still struggle. Standardization, not personalization, has become the norm. America still has a 14:1 teacher to student ratio and universal curriculums are still forced upon each student. But as all parents know, every child is different. And for that reason alone, it might be time for some change.

             In the 1970s, this demand for change evolved into a better understanding of how students interact with academic material. This ‘learning styles theory,’ a branch of educational psychology that – simply put – treats students as individuals, implies that different students use different learning styles to approach different academic situations. For example, while student A might respond well to authority and enjoy knowing what is expected of them, student B might embrace a more unstructured setting with more freedom and uncertainty. Student A might prefer reading a chapter in a textbook instead of listening to a lecture, and student B might prefer neither. Whatever the case may be, it’s important to recognize that if those two students are in the same class, at least one of them is being left behind.

            As the promise of personalized education began to grow in popularity, different researchers started branching off and adopting their own practices, creating over seventy separate learning style theories, all backed by various professors and educators around the globe.

Learning Styles and Virtual Learning Are the Perfect Match            While this expansion may appear promising, it instead brought a devastating blow to the reputation of the learning style experiment. Reviews came flying in, criticizing the theories for inconsistencies, unreliability, and – above all – decentralization. Without a clear leading theory, different supporters couldn’t find practically any common ground, leading to a 1990 ASCD article that compared their struggles to the fable of five blind men, each with a part of an elephant, but none with a full understanding of the whole animal. This expansion of criticism came to a peak in 2009 when Pashler et al. published a paper for the APS that seemed to all but put an end to the learning styles discussion. Since then, dozens of different analyses and publications have cited the 2009 article as the nail in the coffin for the once promising educational trend. Without much academic support, little has changed in the past decade.

            But there’s a lot that isn’t mentioned about the influential 2009 publication.

            Not only do Pashler et al. only study a specific type of learning style theory (preference-based) that does not accurately represent the entire field, but they also spend much of the publication simply declaring that past studies do not prove the validity of learning styles due to questionable methodologies. As of the writing of this article, no publication without deafening criticism has come close to proving or disproving the theory as a whole. In fact, the 2009 article encourages further testing and even outlines the process to prove the theory in just its third paragraph. According to those instructions and the failures of past studies, the next wave of research must include wide swaths of data and educational outcomes never before used in learning style research – a field that was largely filled in decades before computers entered the classroom.

            While, yes, some colorful online quizzes that ask, “What type of learner are you??” and the previously respected “right brain vs. left brain” theory (which is similar to learning styles) have both been practically debunked, the vast majority of learning style theories are still up for academic debate. Proper utilization of technology can answer past questions about the theory and test for its validity. In the right hands, it could personalize learning for students around the world, inspire otherwise disinterested students, and change learning for the better. And while the mystery of learning styles may continue for years to come, the debate should not come to a close. With so much academic potential at stake, we owe it to the young and future generations to give them the best educational experience possible, to build the system around their needs – not the other way around – and to put the elephant back together again.

            The Euclid School was created to test and expound upon these theories. While traditional classrooms or limited studies can only scratch the surface, a digital platform-driven solution that rapidly tests thousands of learning style theories and iterations can truly give insights into how learners learn. Our mission is to deliver a custom-tailored education to each and every student that steps in our digital classroom.

While Euclid School was only founded in late 2020, our team of data scientists have already built a cutting-edge machine learning platform that will enter beta testing with a limited cohort focused on Algebra 1 in January 2021. If you’re interested in joining this free beta program, sign up today, and we’ll contact you to arrange a white glove experience.