Learning in Depth: a double post-mortem part 2

Online vs in-class learning

A few months back I finished teaching two separate cohorts of Learning in Depth in two very different contexts. Part 1 of this series discusses an after-school cohort I taught at a Hong Kong primary school. Today I’ll discuss the larger and more intensive cohort I taught online and where LiD is going next.

If you’re not sure what LiD is, here’s the short version: a project-based learning program in which students are randomly-assigned an everyday topic that they are expected to learn about using Kieran Egan’s cognitive tools. The twin goals are for students to achieve deep knowledge of and appreciation for their topic, and also to learn research methods that can help them bring any new topic to life. If you’re curious, I’ve written more about LiD here and here.

This was the big one, significantly more ambitious than the Inaugural cohort last summer or the program I ran last year in school: it was an eight month long learning program developed by Brandon Hendrickson and Dr. Alessandro Gelmi and mentored by myself and Shelley Smith. We delivered it right here on Substack (posts are gated).

The reasoning behind expanding the LiD cohort from 6 weeks to 8 months was twofold: First, many students (myself included) felt that the inaugural cohort felt like a mad dash, and second, Kieran Egan’s original plan had been to run LiD for 18 years, so it felt natural to extend things longer.

The Course

As for the content, we preserved much of what worked well in the original cohort. ‘Pokes’ were short exploration activities designed to introduce the wonder of a topic and ‘Quests’ were larger, more complex tasks that required a bit more knowledge and input from the participants. We spread six weeks of daily content over roughly six 5-week blocks, spacing the pokes out to just Mondays and Wednesdays, with a quest at the end of each block.

To this core structure, Shelley and I added a series of posts on Fridays alternating between questions posts and source posts designed to be optional and supplemental to the main course content. The questions posts used questioning techniques designed to help students reflect on the process of LiD and on their growing understanding of their topics. Source posts introduced resources and media like podcasts, documentaries, poems, artwork, etc. that the learners might want to incorporate into their exploration, with tips on how to find the true gems.

The blocks were very loosely organized around Egan’s cognitive toolkits: the Somatic, Mythic, Romantic and Philosophic understandings. Without going too deep into the theory underpinning this, cognitive tools are ideas and modes of thinking people use for learning, sensemaking and understanding the world. Some examples include organizing ideas in a narrative structure, using metaphors, telling jokes, using visualization and many other techniques. Cognitive toolkits are collections of cognitive tools that tend to emerge together due to patterns of language use and social interactions. Non-verbal communication corresponds with Egan’s Somatic tools, verbal communication corresponds with Mythic tools, literacy with Romantic tools and academic literacy with Philosophic understanding.¹

Aside from theoretical organization, within the blocks each poke was mostly independent of the others. However, we did try to group activities and sources together to give a slightly different character to each block and to help build toward the quest activity at the end.

Shelley Smith and I were mentors for the program, which put us in charge of delivering the material, helping to guide the students and dealing with any problems with the course. In addition, we built on the foundations created by Brandon and Alessandro, re-writing and adding to course material in order to expand the course from a summer intensive program to fit nearly an entire academic year.

The Crash

Running the cohort was a bit like getting the keys to a Lamborghini when you’ve just graduated from driver’s school.² We had to learn quickly, but that said, the cohort started off really well. Student enthusiasm and participation were both high for the first two blocks and by then Shelley and I had gotten a good feel for running everything. That’s when the wheels of our metaphorical Lambo began to fall off. The wheels being the participants in our cohort. It was late November / Early December and we were nearing the end of the second block (roughly 1/3 of the way through the cohort, about 10 weeks in). Student participation started dwindling and never fully recovered. At the beginning of each subsequent block there was a bump in participation which would fade each time. Eventually the core group of students stabilized at about a quarter or a fifth of the peak response rate and they remained throughout the remainder of the course.

Alessandro, Shelley and I came up with some changes to the overall structure of LiD (more on these below). By the time we added them, it was quite late in the course and they didn’t affect participation much. If my school-based LiD course failed because students didn’t engage deeply with the topic and they were not transformed in the process, the online course failed because we couldn’t sustain the engagement - though the transformation was real.

The silver lining

There’s a point of view where nothing is a failure as long as you learn something - and boy, do I feel like I learned a lot from my many mistakes. That’s not the point I’m trying to make here. I think that this cohort succeeded in a number of ways that my school-based cohort didn’t.

First, the quality of the students’ work over the course of the cohort was consistently good to great. I was really impressed with the ideas, connections and creative work the students were able to generate. Whether they were writing song lyrics or creating maps or timelines or uncovering scholarly debates about their topics they were all thoroughly imbued with meaning. I am still struck and a little disheartened by the contrast between the quality of these LiD pokes and quests and the typical schoolwork responses I see each day because it shows just how little of schoolchildren’s hearts are in their daily assignments. By the end of our LiD cohort some of our students presented a masterwork, a summation of their learning during the marathon cohort and it was beautiful- a profound statement of what is possible that makes me feel incredibly proud to have been a part of it. Although I can’t present those masterworks here- the students were video recorded and don’t want their images to be shared online- you can get a sense of what this type of presentation might look like in a recent open house video.³

In addition to doing great work, the learners really seemed to appreciate the experience. They frequently reported insights that helped them see their topic in a new light, opening them up to new perspectives. Even the students who stopped participating typically expressed regret that they didn’t have time to pursue LiD further. I think this was most apparent in the discussions we had during meetups, which contained passionate discussions about the value of school or the place of subjects in education.

Diagnosing the Cause of Death

I can’t say that everyone’s experience was a shiny ball of wonder- we don’t have feedback from all of the participants and surely some withheld all sorts of valid criticisms. However, putting the feedback we do have together with the dropoff in participation I think we can diagnose what went wrong.

First of all, our 8-month LiD cohort was simply too long. Perhaps this may come across as ironic to those of you who know anything about Kieran Egan’s conception of Learning in Depth because he suggested that students do it for 18 years. Students started with a lot of enthusiasm but it petered out around 2-3 months as the realities of schooling and everything else in life began to wear on them.

Given the fact that the course was so long, the intensity of the demands was too high. We presented the students with three activities a week on Monday, Wednesday and a less-intense optional Friday task. Although predictable, we heard from lots of students who said it was difficult to keep up. Even though the course was structured such that the tasks functioned independently, people seemed to want to do every activity in order rather than jumping to the most recent pokes. Thus, once students got behind, they often fell further and further back.

This brings me to another problem, the overall structure. When our version LiD was originally designed by Brandon and Alessandro, students would do an activity every day leading to a more involved ‘Quest’ on Friday. The weeks were lightly themed to fit roughly according to Egan’s kinds of understanding. For a short, 6-week course that was organization enough, but with our 8-month marathon LiD we could have used a more dynamic structure. The month-long blocks with their same structure began to feel a little same-y and I don’t think the structure lent itself best to deep understanding because the kinds interactions didn’t vary enough with student knowledge.

Ultimately, there are two kinds of failure modes with LiD - in one ‘shallow’ case, students fail to develop deep knowledge and mastery of their topics because they either aren’t challenged or lack access to sufficient, meaningful knowledge. In the second ‘attrition’ case, students are so challenged that they simply can’t keep up or the challenges makes alternative uses of their time more appealing. If anything, our online LiD course has failed much more on the attrition case and not at all due to shallowness. This was particularly relevant because we were running the course for students from age 9 to adult. We needed a better way to scaffold the content and meet students at their level and allow them to engage less deeply when conflicts or difficulties inevitably arose.⁴

Hear the good news: LiD is Dead. Welcome to LiD 6.0!

Yes, the Learning in Depth cohort that I announced at the end of my last post is now underway and nearing the halfway point. The reason that this post has taken three months rather than a week (as promised) is that Shelley and I have been hard at work behind the scenes getting everything ready for this newest iteration.

I have good news to report- things are going much better this time around! Let’s examine why:

First things first, we’ve shortened the course from 8 months to 12 weeks. This timeframe is about as short as we can make it and still develop a relatively complex understanding of their topics. Regarding the structure, Alessandro helped us identify a key feature of Egan pedagogy that’s a perfect for LiD- narrative. We divided LiD into four phases: orientation, complication, transformation and integration. Each of these has a distinct purpose and a separate cadence intended to help students move from early ignorance toward an increasingly independent and internalized knowledge of their topics.⁵ The best part is that this kind of narrative structure works on multiple levels: pedagogically, it scaffolds instruction, gradually releasing responsibility and yet it is far more engaging because it brings a dynamism to LiD that previous iterations lacked.

Given that students most often seem to fall out of LiD due to its demanding nature, we’ve tried to add more granularity and scaffolding to the individual tasks themselves. The course is now on Google Classroom, which allows students to interact much more flexibly than our old interface on Substack. To reduce the cognitive load and time burden we’ve tried to implement different levels at which students can interact with each activity. Pokes have been distilled down to simple questions and students can complete them simply by answering the question. Each poke also has a detailed guide that provides step-by-step instructions, often allowing for extension when students really want to go deep.

For longer-form assignments students now have a logbook. This document acts as both a scratchpad and portfolio for students to keep all of the information related to their tasks. Everything is searchable, so that when students do research or create their final masterpieces they’ll have all of their previous sources to draw on. Another way we’ve tried to make things more accessible is by providing weekly video introductions and having two online meetups per week, providing lots of chances to ask questions and for us mentors to suggest ways of moving forward.

What’s the Prognosis?

We’re six weeks into the current cohort and the patient is alive and well. We’ve enrolled two cohorts, including an open cohort and a dedicated cohort in association with the Davidson Institute’s Young Scholars Program. Our community is as vibrant as I’ve seen it thanks to the online meetups and lots of students who have taken it upon themselves to comment on each others’ work. The cohort is now finishing up 3 weeks of self-directed research. Next we’ll move into transformation, an opportunity for the students to share what they’ve learned over the past 6-weeks but with a creative twist that should help to generate new perspectives.

We’ve made a lot of progress but there’s still so much more we can do. There is still a slow dropoff in student participation but it’s been less extreme than in either previous iteration. Both cohorts have maintained an active core of highly interactive students that’s doing a tremendous job. I’m genuinely excited to follow the students’ progress and I can’t wait to share some of their explorations here in the future. If you’re interested in learning more about LiD or following along with future cohorts, you can check out our website at learningindepth.quest.

1. Phew… that’s about as short as I can get this explanation. Also, I’m not forgetting Ironic understanding - we just didn’t cover it as an entire block.

2. Perhaps, given the fact that Shelley and I both have been in and around education for years, it’s more like getting the keys to Lamborghini when you’re used to driving a bus. Or perhaps it’s more like teaching someone else to drive a Lamborghini when you’ve just learned to drive one yourself.

3. This is my own pale imitation of the students’ far better work. I do plan to gather more to share in the future.

4. A better way than simply declaring content as optional and talking them through the procedures in a blog post format.

5. To be fair, students enter with a range of awareness of their topics - some start with relatively broad understanding.