“Adaptive learning”makes sense because it is designed “to orchestrate the allocation of human and mediated resources according to the unique needs of each learner.” However, adaptive learning systems are designed to allocate a meager set of resources from a narrow pool generally developed in house, ignoring the dynamic array of open educational resources available online, without granting learners the freedom to make decisions, based on system-generated recommendations. There are a lot of good leaning materials out there, but learners are left to sift through them, and the results are inefficiency at best and perhaps even frustration. So…
To solve this problem, let’s build “CELLO,” the “Curation Engine for Locating Learning Objects.”
CELLO will allow educators to offer collections of learning objects designed to help learners meet well-written learning outcomes (AKA “objectives”). In other words, the content of what is to be learned is determined by educators in collaboration with employers, who define a desired competency. They then develop a set of learning outcomes that lead to competency. The program is arranged as a series of learning outcomes, learning opportunities, and assessments, allowing modular “personalized pathways” to competency respecting what experienced learners already know and can do. This approach promotes “learner agency,” the power for learners to make important decisions about when they engage with instruction and when they skip to the assessment, as illustrated in the “Progress & Navigation Dashboard” below. The result is efficiency — effective learning tailored to individuals, in less time.
For each outcome, a curated set of learning objects, a “CELLO Set,” is created. To create these sets, the instructional designers and content experts review the available materials designed to meet that outcome (videos, websites, simulations, documents, etc.) and create “Amazon.com-like” descriptions of each they deem to be capable of leading to mastery of the learning outcome, and enter them into a database.
When a learner sets out to master a learning outcome, the CELLO Set is presented, pre-sorted based on the learning objects’ history of success “with learners like you” (another practice borrowed from online commerce). However, determining “learners like you” is no small task, as many factors might be considered, and a complex profile could be used to make these determinations. The goal will be to increase the sophistication of learner modeling over time, but for now, we propose beginning with a simple assessment, by the learner, of how difficult it might be to achieve the learning outcome. (See below.)
After selecting a learning outcome and rating the perceived difficulty, the student is offered a curated list of options, sorted based on previous learners’ data, using a “success with students like me” approach based on perceived level of difficulty. Learners are free to re-sort the list, to see options in terms of past students’ ratings of quality, efficiency, or effectiveness.
After selecting and using a learning object, the student quickly rates it on perceived effectiveness, efficiency, and quality, and following the assessment the system records a new database entry, upon which future curated lists will be presented.
After each student/object interaction, the system becomes “Smarter” — better able to make recommendations, with no human intervention needed. CELLO can personalize learning.
Of course, CELLO isn’t always the right instrument for the job. Cello is a great option for developing knowledge- and comprehension-level outcomes, but other methods will be best suited to develop certain higher-order skills.
Let’s build CELLO, to personalize the acquisition of entry-level knowledge and comprehension, saving time and money for the development of high levels of competency.