1.0 – Reflective Learning in Context
This essay first highlights some learning mechanisms and teaching strategies, to set the table for our teaching options. We then focus on Reflective Learning and its attributes and advantages, while leaving other topics open for future discussion.
Learning Modes. From fact-memorization (ala Quizlet) to metacognition (ala Tanner, 2012), there is a wonderful diversity of student mental activities that teachers (including professors and tutors) seek to spark, shape and invoke. Because boundaries between types of learning activities (e.g. active, reflective and adaptive) are fuzzy, and perhaps a tad arbitrary, it becomes important to establish some definitions. Teachers can indeed engage students in an endless variety of learning experiences including flipped classrooms, project-based learning and more traditional platform-style lectures. And any of these could incorporate a variety of distinct “learning modes”. For example, while working on a new project one might learn some vocabulary, discuss the meaning of new terms, create a concept-map to frame this new knowledge architecture, and write about a curious or challenging concept within this domain. For the sake of discussion, we posit here 4 learning modes (each of which implies a distinct teaching strategy):
traditional = memorize facts, aka “rote learning” but facts are VERY important for other modes
active = doing something e.g. build something or explain a concept to a classmate
reflective = more introspective, mental activity and can segue into
adaptive = algorithmic instruction derived from machine-learning tools and big data
While not exhaustive, these 4 modes encompass much of what students are doing both online and in the classroom. This conclusion depends, of course, on how one interprets each mode. For example, Active Learning normally means something more active than taking notes and would naturally include collaborative and project-based learning. For parsimony, it is often preferable to associate a given learning activity with a specific mode. In the case of Pair and Share (where students sitting next to one another in class briefly discuss a question posed by the teacher), this is a fun, easy-to-implement and much more Active Learning activity than rote listening. As a student ponders how to explain “tyranny” to her classmate, she might well reflect on different examples and their virtues as she formulates her explanation, but we’d lean towards Active as its home category, if forced to make a choice. For Reflective Learning a better example would be the One-Minute Essay that some teachers ask students to write at the end of lectures. Here students must consider i.e. reflect upon what they have learned and then must put something of the lecture in their own words, which might be viewed as a deeper level of reflection or use of brain networks.
As the foregoing illustrates, there is an unavoidable fuzziness to the boundaries between learning modes, but these modes at least offer a starting point to examine the essential elements of learning. One might look to learned academics and/or high-rankings in Google Scholar to better glean the true learning modes at work today, but a third arbiter is the brain itself. In other words, we should consider the elements of different learning modes, how they differ from one another and, most crucially, how they engage brain systems. If learning modes have distinct effects on learners, it is because each impacts the neural storage/organization of knowledge and skills in distinct ways. This is the crux of MazeFire’s approach to learning and EdTech: we hope to understand how each learning mode works at the neuronal-network level and what this might tell us about each strategy’s strengths and limitations. The parsing of modes is not to detract from any one mode, but rather to highlight what each mode is doing in our brain and how the available learning modes and related resources can spur cognitive advancement.
Reflective Learning
What is Reflective Learning? All learning modes entail delivery of facts, knowledge and experiences into neocortex, the folded layers of cerebral cortex that make up most of our forebrain.
Rote learning, via traditional instruction, often entails the pipelining of facts into the student brain and might be viewed as a necessary evil: it is not easy to learn the rules of French grammar if you’ve learned none of the words. Nor can we sort animals into phyla if we know only their names—we’d also need to know essential characteristics. But consider if you are asked about whales, about which group of animals to assign them to. This requires something beyond storing and retrieving bits of information. We know whales live in the water and are well adapted to swimming and aquatic feeding, thus making “fishes” a natural category for them. But if provided more information about e.g. brain structure, reproduction, circulatory and respiratory systems, it eventually becomes clear that whales belong with the mammals and not the fishes. While students could just memorize which animals are fishes and which are mammals, the process of thinking about criterion for different classes, retrieving stored information, applying newly acquired information and making conclusions is (no pun intended) a different kind of animal: it is a reflective process.
At the level of neuronal computations, it is far from clear what is going on in reflective learning processes, but it is inherently recursive. (This also applies to machine learning/AI and the June 2017 Scientific American has an article on the latest bottom-up vs. top-down learning approaches which are essential to the advancement of AI). One core aspect of recursion is reciprocal (including repeated) interactions, and this occurs in all these domains. Indeed recursion is touted to be a defining feature of human language (Hauser et al. 2002) and as such is indispensable for the highest levels of human cognitive activity. Reciprocal neuronal interactions epitomize the architectures of our brains, from the neuron-to-neuron level, up to the highest processing scales, including e.g. gamma-band EEG rhythms which underlie conscious percepts. While the exact linkages between recursive neuronal operations and reflective thinking remain to be deciphered, what is certain is that there are many recursive pathways between stored representations of e.g. fishes, whales and clades, and that these become engaged when e.g. one is seeking to assign whales to the correct group or clade of animals. More generally, the very act of thinking entails massively-parallel, recursive processes acting at many scales of reciprocal neuronal/network interactions.
In a less arcane vein, we might like to list reflective activities and could include such things as thinking about what you do and do not know and trying to “figure things out” so as to e.g. assign an animal to a biological clade or solve a Digital Maze puzzle. There is clearly an inductive, problem-solving aspect to such activities, which generally use both bottom-up processing of incoming sensory streams, as well as neocortical “top-down” information processing. There is certainly an active component here, but the more internal nature of mental deliberations is intrinsically reflective. Just by having students write about things they just heard in lecture, or by asking them to summarize the high points of the week’s course content, we are asking them to retrieve and recombine information and to export it in a new form, i.e. we are asking them to reflect. An especially germane activity is “Muddiest Point” reflections, where students write down the one point from the lecture that most confused them (Krause et al., 2014), which as a bonus provides valuable feedback to the teacher. Such activities fit nicely with other learning strategies such as spaced repetition and provision of motivational rewards such as extra credit or badges.
Reflective Learning is thus a distinct class of learning activity that both relies upon recursive neuronal operations in the cerebral cortex (including both neocortex and hippocampus) and that produces new levels of understanding as well as student insight into their state of learning/knowing. This kind of cognitive outcome is expected, based upon first principles of synaptic learning theory (to be discussed in our Neurobiology Primer), due to the reinforcement of extant connections, while also increasing the numbers and kinds of new connections to discrete information nodes in neocortex. Such activity can build student confidence and help them transition from wondering about to knowing the specific topics at hand. Reflection is also integral to yet higher-level learning activities where e.g. we encourage students to think about how they are learning and what the best strategies would be. Termed metacognition, this is a trending topic in undergraduate education (see e.g. Tanner, 2012). Finally, from both a cognitive and EdTech standpoint, we are interested in the relative merits of reflective vs. adaptive learning ala Knewton (Conklin, 2016; surface), and we anticipate the release of a companion essay on that this particular comparison in the near future.
