2017 is the year of micro learning. It took 14 years since our initial work in 2003 until the concept hit mainstream. During the first half of this year, I have read several blog posts and tweets since the start of this your and one thing struck me: The focus on learning resources and chunking them into digestible sizes. It appears that video is the big thing in micro learning these days, while it is not.

This post is a response to Mirjam Neelen and Paul Kirschner's post that comes with a lot of references but leaves out the most important aspects of micro learning and argues that micro learning is a meaningless concept. One thing that the authors got wrong is the honorary reference to micro instruction from the 70's. Back then it was called micro teaching and not micro learning for a reason. Mainly, micro teaching is about presenting the teachers ideas and concepts and not about the student's learning and performance. More importantly, micro learning is not content-centric just-in-time learning, as the article (like many others) puts it.

I agree with the Mirjam and Paul that micro learning is not the solution to all learning in the workplace. However, this is true for all concepts of learning. The presented conclusion is also misguided based on the selection of definitions. Particularly, because the authors stuck with hermeneutical reflections, which try to fit practically everything into the concept. I agree that this is not helpful, but leaving out those who operationalise the concept, bring it to practice, and integrate it with other approaches of technology enhanced learning is equally unhelpful.

One aspect of micro learning is spacing. Spacing refers to learning activities that distribute learning activities throughout time. This is a really old concept and dates back to the beginnings of modern educational psychology (Ebbinghaus, 1885). In a nutshell it says, if learners repeat learning activities after their initial attempt, then forgetting takes longer and retention is improved. You will find that many articles on micro learning repeat the obvious: Byte sized instructional chunks are easier to spread across time than long lectures that require a reasonable amount of time to digest. This has lead to minimising learning resources - most notably videos - into learning nuggets. This idea of learning nuggets is not entirely new, either. However, just focussing on learning nuggets don't bring you anywhere with micro learning, but it leaves you in the field of micro instruction.

In 2003 we started from a completely different angle: motivation. The initial question back then was: how can we motivate learners to engage with learning activities in order to increase their time on learning? (or better: this is how I would phrase it today). This question came from a problem of learning in the workplace that is familiar to many: conventional learning solutions require too much time and get in the way of normal business activities. The effect is that learning is not happening, discontinued, or insufficient time is dedicated to learning. Often learners in businesses have long gaps of no-learning between their learning episodes. As a consequence much repetition had to be done before the learners could continue on their learning journey. The goal of micro learning is to increase the time on learning and thus reducing the time needed for getting started with new more complex learning activities.

The initial concept we came up with was a brute force approach to extrinsic motivation by placing learning activities in front of activities or resources that the learners want to reach, like unlocking their screen from the screen safer. This is similar to von Ahn's Captchas and the concept of human computation from around the same time (interestingly, micro learning concepts appear in von Ahn's duolingo, much later).

In order to get started, we used learning flash-cards and the Leitner-system as micro content for our first prototype. Obviously, individual flash-cards are quickly to perform and they are not creating frustrating experiences when one or two have to be handled before getting access to other activities.

One drawback with flash-cards is that their conventional form provides only self-reported evidence on learning. This does not always match actual achievement. Moreover, in the workplace self-reported evidence is not enough: many professions need to provide evidence of learning and understanding for compliance reasons. Consequently, organisations and learners need to know whether actual learning took place. This brings to us to the third ingredient of micro learning: performance.

We operationalised these concepts by introducing multiple choice test questions instead of flash cards. This allows using actual performance instead of self-reported statements for gaining evidence on learning. The criticism was that this is just behaviouristic learning using drill and practice. In the retrospective, using multiple choice questions at the height of "constructivist e-learning" was a bad choice, but it yielded better results on the level of achievement than self-reports on conventional flash cards.

The key challenge for micro learning is not its potential but its operationalisation. The two concepts motivation and performance are integrated in self-regulated learning. Self-regulated learning means that learners use provided information to assess and control their learning processes. Indeed, there is some student behaviour involved, but this is not the same as stimulus-response. From this perspective, learning activities can be broken down into four components:

• Tasks and Cues,
• Actions,
• Feedback, and
• Reflection

These components are the core of what we all know as feedback loops. According to Butler & Winne, performance and feedback are tightly coupled. Artificially separating the feedback from the learners' actions reduces the effectiveness of the feedback, because the learners find it more difficult to relate the outcomes of their actions and strategies with the feedback they receive. The quality of learners' reflection on and decision making for their learning depends on their ability to link their actions and the feedback they receive on them. The availability of such feedback influences the motivation of learners - or in bad cases their demotivation and drop out. Micro learning solutions typically include immediate feedback after each learner activity based on the learner's performance.

Minimizing learning activities means that further reduction or transformation of an activity would break the feedback loop by removing one or more core components. This is why the "learning" is different from the "teaching" in the older concepts, that do not include learner actions or feedback.

A second criterium for micro learning activities is the independence of activities. This means that two activities are not or at best loosely coupled. Therefore, learners can interrupt a their learning after completing any activity. This implies the ability for entering the learning process without having to repeat other activities.

In the work on the Mobler App we reduced the learning activities into micro feedback loops that can be completed in under a minute. This lowers the barrier for engaging with learning activities. In a workplace setting such barriers play a crucial role for the learners' time-on-learning: If a learner can quickly start and complete a learning activity, they find it easier to use in-between settings for their learning.

Reducing the entry barrier for engaging with learning can be quite hard with conventional learning management system, because these systems come with a lot of overhead before the learners actually spend time on learning. This includes authentication, finding or selecting courses, and navigating a course structure. In the way to the learning activity there are so many distractors that these side activities can easily become the entire activity.

Most micro learning solutions go beyond bite-sized learning activities and also remove hurdles, lower entry barriers, or push learners with reminders and notifications. This creates learning opportunities over coffee breaks, between meetings, or during commute. In these settings, it is not the single activity that makes the learning just like reading a page of a book does not make the reading experience for the entire book.

When it becomes easier for learners to engage in learning, distributing learning activities across time becomes also easier. Eventually, spacing appears as a side effect of micro learning rather than at the core of the concept. The distribution of learning activities across time happens because learners find it easy to engage and to disengage with learning activities. If pushed, learners are more likely to engage in a few micro activities because they know that they are not stuck in a lengthy process and feel in control of the time they dedicate to learning.

In addition to the core affordances of micro learning selection processes guide the orchestrating of the learning activities. With flash-cards this selection is structured by the correctness of previous answers. Such selection ensures that activities are not appearing too often. Depending on the learning performance this also creates a spacing effect for the individual learning activities.

A third aspect of spacing in micro learning is repetition. As bite-sized activities allow learners to engage and disengage quickly into learning, repetition allows verifying performance and work against forgetting. Repetition can space learning as well because some questions tend to be harder, are easier forgotten than others and thus need to be repeated more frequently.

Micro learning emphasizes in-between learning and can benefit from the affordances of modern devices. However, micro learning and mobile learning are very different concepts. Micro learning is primarily about the structure and the arrangement of learning activities as feedback loops. Motivational aspects for engaging in learning are more important for micro learning than the mobility of the learners. This does not mean micro learning on mobile devices do not help learners to expand their learning environment into new and unusual contexts. Our results indicate that this kind of contextualisation seems to more related to reduced barriers for some micro learning activities. These reduced barriers account more to the ubiquity of mobile devices than to potential mobile characteristics of micro learning.

Our practice indicates that micro learning has its place, both, in workplace learning and in higher education. The concepts of micro learning are useful to enrich the learning experiences and broaden the learning environment where conventional macro learning solutions are unsuitable.

Learners frequently report that extending more conventional educational approaches with micro learning activities helps them to spend more time on learning and to learn more continuously. However, typically our participants disagree that the micro learning elements are suitable replacements for similar educational tools and concepts. Although this appears discouraging, the disagreement goes in both directions.

Understanding micro learning as minimal indipendent feedback loops also helps to reflect learning designs beyond the macroscopic level and focus on arranging and integrating such activities. I found the concept also useful for reflecting and operationalising learning analytics for orchestrating learning.

While micro learning appears to be trendy in 2017, I agree that labelling everything as micro learning is wrong. Because it is hard to tell micro learning solutions apart from those that try to hitch a ride, I offer five questions that you can ask when looking at self-proclaimed micro learning solutions.