The Blurry Definitions of Adaptive vs. Personalized Learning

Educators still use "adaptive" and "personalized" learning interchangeably. Does it matter?

In June of 2015, leaders in adaptive learning hashed out the definitions of personalized and adaptive learning at a summit in Santa Fe, NM, hosted by WCET (the Western Interstate Commission for Higher Education's Cooperative for Educational Technologies). And now, more than a year later, the adaptive learning community has moved on. The terms have been defined: "personalized learning" is any customization of learning by an instructor, while "adaptive" refers to technology that monitors student progress in a course and uses that data to modify instruction in real time.

The formal discussion of what those terms mean, at least among experts, is over.

"Today, it's not even part of the conversation," said Niki Bray, WCET adaptive learning fellow and an assistant professor and instructional designer with the University of Memphis School of Health Sciences. "We're not even having that discussion with the vendors anymore."

But despite the fact that adaptive learning experts have settled the definition question, many rank-and-file educators still use the terms "personalized" and "adaptive" interchangeably.

"I think there is a little bit of blurriness there," said Martin Kurzweil of Ithaka S+R, a not-for-profit group tasked with studying technology in education. "I think one of the reasons there is some confusion is that some authorities in the space have their own preferred terminology, and use that terminology to describe whatever it is they're talking about."

Tracing the Terminology

When it comes to personalized and adaptive learning, no matter where you dig, you'll eventually strike the Gates Foundation. Since 2013, the foundation has been funding postsecondary personalized learning initiatives as part of its mandate to increase college completion rates and ensure all students have access to a high-quality education.

Because the Gates Foundation has fostered the personalized learning market, it makes sense that the organization has also influenced the language being used to describe that work. Gates uses "personalized learning" in its literature, and sees adaptive as a tool to personalize instruction. As more organizations try for Gates grants, that version of the terminology has spread, but not all the Gates grantees are using Gates' language with the same precision.

"Some of them make very little distinction between adaptive learning and personalized learning," said Michael Feldstein, a partner at educational technology consulting firm MindWires. Feldstein and his partner Phil Hill are the unofficial guardians of the conversation about the semantics of personalized and adaptive learning. They have written extensively on their blog, e-Literate, about the need for a clear definition of both terms.

Feldstein argues that personalized learning has been used as a marketing term rather than as an academic one. That, he says, is where the real danger of muddied terminology lies — personalized learning is often reduced to "adaptive plus some kind of vague, new and improved label."

His concern is that the marketplace, and its language, is exerting pressure on teaching — something he refers to as the "solutionizing" of education. College instructors, many of whom have no instructional training, may confuse the benefits of personalized learning (a style of instruction) with adaptive technology (a product or solution). Feldstein's concern is that some institutions might take the easy way out, choosing to buy products rather than change teaching methods to include a personalized approach to learning. That, he said, might be harmful to students.

"Our position is there's very little evidence that adopting these technologies without adopting the teaching practices that these products are meant to support has broad impact on student outcomes," said Feldstein. "If we know that teaching practices have the biggest impact overall on outcomes and we keep confusing products with practices, we're not going to get anywhere in terms of influencing those outcomes."

The conversation about what to call adaptive technologies started almost as soon as the first Gates grants were awarded, as the higher education community repeatedly asked itself what personalized learning is, and how to define it.

Adam Newman of Tyton Partners, an education-focused investment banking and strategic consulting firm, defined personalized and adaptive learning for Campus Technology in 2014. Personalized learning, he said, is an umbrella term encompassing many different approaches to customizing instruction. Adaptive learning falls under that umbrella, along with other practices like competency-based education and differentiated instruction.

Newman says that definition has changed very little in two years. "Adaptive has increasingly become more about the features and functionality of the solutions as opposed to a style of learning in and of itself," he said. "I think terms like 'personalized learning' and 'student-centered' are more holistically centered on the approach or the practice. I think adaptive is increasingly seen as styles or tools to apply."

The University of Memphis's Bray defines adaptive learning as being computer-mediated, using algorithms to adjust in real time to student progress, and delivering big data to instructors and institutions. "There's no face-to-face teacher who can do that," she said. "You can't offer adaptive learning face-to-face without some level of computer mediation."

Definitions in Context

Kurzweil doesn't see the muddiness in the terminology to be especially damaging. "I'm not sure it makes a difference at this point," he said. "I think typically the advocates for and practitioners of personalized learning understand the distinction between the more general personalized learning concept and adaptive learning technology, but they refer to everything as 'personalized learning' because that's the lens through which they view it."

Newman is likewise undisturbed. Although he believes the definition of adaptive learning has finally settled, Newman isn't sure the new terminology of digital learning will ever harden into a standard glossary. "I think some of these terms are always going to have multiple definitions underneath them, depending on who you are and how you're using them," he said.

Multifaceted terms that mean different things to different people shouldn't be a problem, added Newman. The best way to achieve clarity is to use specific examples when discussing them to ground all stakeholders in the same context. "Understanding that these terms can be slippery, that these terms can have different connotative definitions or meanings, we've got to make sure we're being specific and that we understand our audience," he said.

Newman says there are different terms that need definitions now: "digital learning," which has emerged recently as a catchall phrase, as well as "open educational resources" (OER) and "courseware."

The interchangeable use of "personalized" and "adaptive" may even be helpful, according to Bray. "Personalized learning," she said, is a less threatening way to introduce concepts like adaptive learning to people outside the adaptive community.

Bray tells a story about a friend at another institution who was introducing adaptive learning on campus and needing to name the department spearheading the initiative. The friend was torn between using the words "adaptive" and "personalized" in the department name, but ultimately settled on "personalized." While "adaptive" might be intimidating to professors who don't understand the concept of adaptive learning, "personalized" is a gateway term that leads people to the concept of adaptive learning, Bray said.

"When I first talk to people who are not very technology-savvy, I do say 'it's a very personalized way of delivering instruction and allowing students to learn' — and then as the conversation goes on, I ease them into the fact that, technically, the term is called adaptive learning," said Bray.

For Bray, the discussion about terminology is wasted energy when there is work to be done. "How do we scale it? How do we design it? Do we have to have instructional designers to do this effectively? What makes for effective design in adaptive? These are the questions we are discussing now," she said.

Feldstein, for one, would like to see standard definitions, recognizable to all educators, emerge. "I think it would enable academic communities to have the conversations they need to have in order to identify and solve their problems," he said. "The problems we have are teaching problems. They're not product problems."

Featured

  • AI-inspired background pattern with geometric shapes and fine lines in muted blue and gray on a dark background

    IBM Releases Granite 3.0 Family of Advanced AI Models

    IBM has introduced its most advanced family of AI models to date, Granite 3.0, at its annual TechXchange event. The new models were developed to provide a combination of performance, flexibility, and autonomy that outperforms or matches similarly sized models from leading providers on a range of benchmarks.

  • Two autonomous AI figures performing tasks in a tech environment; one interacts with floating holographic screens, while the other manipulates digital components

    Agentic AI Named Top Tech Trend for 2025

    Agentic AI will be the top tech trend for 2025, according to research firm Gartner. The term describes autonomous machine "agents" that move beyond query-and-response generative chatbots to do enterprise-related tasks without human guidance.

  • minimalist person icons connected by straight lines and circles

    Microsoft Revamps Loop Collaboration Platform

    Microsoft Loop, the online collaborative platform in Microsoft 365, is getting a number of new features and an overall redesign.

  • AI-themed background with sparse circuit lines and minimal geometric shapes

    New Copilot Studio Feature to Introduce AI Agent Building Tools

    Microsoft has announced plans to roll out a public preview of a new feature within Copilot Studio, allowing users to create autonomous AI "agents" designed to handle routine tasks.