Everything is upside-down: turning lectures into homework with problem-based learning

The other day, I stumbled (via Tony Baldasaro) on this gem:

How much more could happen in our classrooms if we created more opportunities for students to learn basic skills and content outside of class? … Class, rather than being a time when all kids sat and received instruction, could be the time when they reinforce skills by doing problem sets, worked on real-world application projects, collaborated with teachers to reinforce concepts, etc…

The post is called Inversions; go read it, it’s only short.

This is such a wonderful, simple idea. And of course, many good instructors and educators are already doing just that — as Chris Lehmann points out, this is essentially what happens in English class when kids read a book as homework, then discuss it during class time. Students use out-of-class time to acquire content, freeing up class time for process. Because processing, doing, is how we learn, and students can get instant feedback from the instructor. Dialogue happens; moreover, students have the opportunity to learn vicariously from other students’ participation.

But this isn’t happening enough in universities, for reasons I have written about before. Big classes and a student body working to pay university fees — or to be able to afford food — mean that often, lectures become an info-dump, because you can’t guarantee that the majority of students have done the reading — and in my view, good teaching takes up from where the student is, not where they should be.

And I do get tired of the sound of my own voice in a two-hour lecture. Oh, I can teach for two hours; this post isn’t coming from a place of laziness. On some level I am probably even a bit of a show-off, or I probably wouldn’t enjoy teaching as much as I do. But, you know, no matter how enthusiastic I am, just talking for two hours is going to lose even the keenest student for periods of time, as their attention ebbs and flows. Estimates of attention span vary wildly, and a big chunk of that is about whether you are in flow.

Passive listening probably does not encourage flow in our students.

Attention span also varies as a function of ability, which is one reason why it’s so important to teach in a way that reaches everyone. And it’s unreasonable, I think, to expect anyone’s attention to last for a two-hour lecture, which is why so many of my colleagues are currently trying to think of ways to break up the time a bit. (The university schedules two-hour lectures in the way that many people schedule one-hour meetings: it seems to be a convenient and universally-understood unit of time, but may not be exactly what is needed.)

So how about we approach this problem from the other direction: make the classroom about practice, and perhaps we can nurture people’s curiosity in the topic and encourage them to pursue the more detailed background content afterwards?

Obviously this strategy is not without risk. Techniques like problem-based learning (PBL) have been found to improve students’ engagement and critical thinking skills, and students who have used PBL seem to hold their own against students educated more traditionally. But I have heard many concerns expressed that PBL can lead to patchy subject knowledge, though I am having trouble digging up much in the way of evidence for that (if you can help me out here, please leave a comment!). Wikipedia has a nice section on the cognitive load issues around problem-based learning; the key thing seems to be to start gently and gradually withdraw support, with the instructor increasingly becoming more of a facilitator.

I wouldn’t necessarily have tried this with first-year students, who perhaps haven’t acquired enough basic subject knowledge. But final-year students have been up to their elbows in the subject for long enough that I figured I could probably meet them halfway.

So, I rewrote my lecture.

In fact, my slides didn’t actually need a great deal of reworking, though I took some more of the text off them. I made lots of duplicate slides: the first with an image, and a question or two; the second, with simple labels. It was a pretty basic format: here’s some stuff — now figure out what you’re looking at.

PBL hippocampus question.png

And then, when they’d had a few minutes, in small groups, to try and work out what was going on, I’d ask for suggestions, and we’d talk a bit about those, and then I’d show them the second slide:

PBL hippocampus answer.png

… and we’d talk about that for a short while. I started off with some basics, and then we got into more and more complex stuff. Occasionally I would remind them, “start with what you already know.” Students had a worksheet that duplicated the images and questions, so they didn’t waste time and attention copying things down, and could concentrate on the what and why.

We did this for two hours (with a break), in a warm lecture theatre, in the afternoon, and nobody fell asleep. Students asked questions, made guesses. It was genuinely interactive.

In many ways, I was lucky. This lecture was all about the visuals: pictures of brains with stuff wrong with them. Had I been discussing highly abstract and theoretical concepts, it may not have worked well. Further, the lecture theatre was pretty much exactly right for the size of class: small, with about 60 seats and an aisle up the middle. I could, and did, reach all the groups; had we been in the 450-seat lecture theatre with people sprinkled everywhere, much of that class dynamic and atmosphere would have been lost.

Of course, not everything went brilliantly. There was a little too much content, and what I should have done was set the remainder as homework, rather than trying to cram it all in. I lost one group at the break, though this isn’t uncommon and you never really know why they’ve left; often it may be nothing to do with you and everything to do with their personal circumstances, and I never like to ask, in case it really is the latter and they are mortified that you’ve brought it up, or noticed their absence.

I won’t really be able to gauge the success of the session until the exam results, and student module evaluations, are in. But overall, it felt right. It felt like a good way to teach, and I really, really hope it inspired students to tackle the background reading. The explicit feedback I have had from students so far has been pretty positive, and a colleague who sat in on the session to observe seemed to really enjoy it, and said some very nice things. All of which gives me a little more faith in my own experience and enjoyment of the session.

Next stop: trying this again, with a bigger class. Anyone want to play along?


Filed under my stuff

14 responses to “Everything is upside-down: turning lectures into homework with problem-based learning

  1. Patchy subject knowledge is a good question. I don’t feel that any of the approaches covered by the umbrellas of ‘informal learning’ or ‘experiential learning’ (or PBL) deal with this at all well.

    Or, at least they don’t when there is a prescribed syllabus or curriculum. I think this is the source of some of the exasperation that seems to permeate the debate.

    I honestly don’t know the answer to this problem and, like you, am looking for answers.

    Daniel Willingham’s video Teaching Reading IS teaching Content reminded me of time spent coaching people to score highly in the essay component of the GMAT exam. I tried every ‘writing skills’ technique I could think of but eventually discovered that the best way to do it was to fling ‘content’ at the learners – not about writing but about the world.

    There were fewer ‘bad writers’ than there were clueless students who didn’t realise the importance of curiosity and general knowledge.

    So, content and cognition are inextricably interlinked. But lecturers have traditionally only taken responsibility for content. Here’s the facts and now you’re on your own (I wonder if THIS is the reason ‘attention’ is such a good predictor of success).

    As you’ve identified, this is a zero-sum situation. If you put yourself on the side of helping students with cognition, you will necessarily have less time for content. And this might lead to patchy subject knowledge.

    Two initial conclusions:

    You can’t do this on your own. If there’s a canon of knowledge and a prescribed set of learning outcomes (ie a curriculum) such teaching techniques can be risky.


    Lecturers have always been happy to absolve themselves of responsibility for content. (Cynical? Yes. True? Mostly.) It’s ‘their’ domain. But there’s no substantive difference cognition and content.

    [Chris, I’m in the middle of writing something on the problem of content and apologise for firehosing/hijacking. This comment may also be even less coherent than my blog posts. Basically, I’m saying we have to learn to let go of content. We should be guides to the domain – not guards.]

    • finiteattentionspan

      Well, first, no apology required! Comment firehosing/thread hijacking is always welcome; this is how real discussions get started.

      I’m tempted to call the whole content/cognition thing a wicked problem, but that seems somewhat defeatist. It comes back, I guess, to the whole “Massification And McDonaldsization of HE” thing: big classes just invite content-dumping, because interactivity and engagement are so hard to sustain and monitor.

      FWIW, I’m with you on being a guide, not a guard — and on substituting thinking skills for content delivery, though I think this should begin much earlier, so by the time people get to university, they are more in touch with how to learn and governing their own motivation and attention. Every time I read about education league tables (schools or universities), a little part of me dies: that approach just serves to entrench the education system in a content-based approach, as people teach to the exam syllabus in a bid to improve their institution’s standing within the league. (In fact, I think schools could do so much better if they followed the skills-based model, but I completely understand why most are too scared to try.)

      Thanks for this; looking forward to reading your post.

  2. One more thing (it appears that this has happened):

    I won’t really be able to gauge the success of the session until the exam results, and student module evaluations, are in

    This bit is the key.

    You don’t have to wait till later – we can now, better than ever, track in real time the ‘content’ learners are working on by encouraging them to develop wiki sites or participate in LMSs/CMSs/Moodles and all the rest of it.

    In the past, you couldn’t very well rummage through their notebooks to see if they were studying the right things. So you had to dictate lecture content – some universities even have compulsory attendance at lectures.

    But now, it is possible to allow some flexibility. You can rummage through their notes to your heart’s content and fill in the gaps later.

    One of the criticisms the old timers level at PBL (and the rest of the learning 2.0 umbrella) is that they’ve seen it all before. And abandoned it. Thus, their criticism is that these are just fancy names for things we should have been doing all along.

    But things are different this time. We have the technology.

    • finiteattentionspan

      You are, as ever, quite right.

      My first response is to tell you all the reasons why I don’t think these ideas would work with this particular class (it’s only a half module; they already have to do a non-assessed piece of coursework which only indirectly relates to this particular lecture; they are strapped for time because of coursework demands in other modules and because of the other constraints on their time, already alluded to; other staff on the module don’t teach like this and my lecture is only one out of five). But that’s all horseshit, really.

      The obvious thing to do with this, next year, is to make changes to the non-assessed coursework. (This needs to go through the proper channels; I can’t just change stuff. But it’s achievable.) Out: paper-based assessment. In: contribution to a blog, or a wiki. There are some issues there about ensuring that the exercise acts as exam preparation, but I doubt that they’re seriously problematic.

      Nice, thank you. These conversations are how progress is made.

      (My favourite Far Side cartoon, by the way.)

  3. To make this work, a major obstacle is how much the students have to unlearn.

    Students, particularly in university settings, are usually there because they’ve gotten very good at listening to lectures. They have often been conditioned — hard — not to interrupt and not to make mistakes. Consequently, they are scared.

    Ask them a question is a setting that is even a little lecture-like, and watch a smart, opinionated person transform into a rabbit in the headlights.

    • finiteattentionspan

      There’s a Mark Twain (? Possibly some Churchillian drift) quote about the function of education being to unlearn as much as possible. And yes, absolutely. Isn’t it sad? :(

      The worst thing is that students who won’t ask questions in a lecture, I guess for fear of ‘looking stupid’, come and ask me afterwards — when I’m more than happy to answer, but it’s such a lost opportunity for the other students, who never get to hear the exchange. Maybe I should log those comments and make them available to all the students. *note to self*

  4. Nifty idea.

    I did English as an undergrad, and we had a (slightly) similar approach. The course had almost no set texts, and lectures were not merely optional, but massively over-scheduled. You physically couldn’t go to absolutely everything that might have been relevant. But, crucially, these weren’t perceived as where the important stuff happened.

    The important stuff was the seminars (groups of 10-15, for a couple of hours, in a lecture/discussion hybrid of the kind you describe) and the “supervisions”, which were weekly, hour-long blocks of 1:1 or 2:1 tuition, where you’d discuss prepared essays, exercises, or reading.

    Now, I’d imagine this is monstrously expensive, and more than a little disruptive to working academics, but you get through a phenomenal amount of material, and really take care to cement understanding. It’s also a lot of fun.

    • finiteattentionspan

      That sounds delightful! Not to mention heavily student-driven, in that you actually had to consider which lectures to attend. Wow. Now I have education envy.

      I absolutely agree with you about the ‘important stuff’, but this is in stark contrast with how we (teaching staff and students) mostly seem to think about things, namely that the lectures are the meat, and the seminars, etc, merely trimmings. I think we have it all backwards.

      You are right about expense, too. lectures are very, very cheap to provide. If we went to a small-groups model, we’d probably have to treble our staff, or cut our student numbers by 2/3. Which is a lot more how it used to be; times have changed.

  5. Been meaning to get here to post for a while (but procrastination rewards in the form of all the interesting comments to read beforehand).

    Lovely example of PBL – thanks for sharing. I do think one of the questions we should ask as learning designers is where can we leave information out (tell less) and let that gap (in the form of questions, etc.) draw the student in. This is a great example of creating the gap for students to fill in, and give them a reason to pay attention to the outside of class reading, etc., knowing they’ll actually need to do something with it in class.

    Made me think of a favorite old resource on PBL: Dan Tries Problem-Based Learning: A Case Study http://www.udel.edu/pbl/dancase3.html – good examples of leaving info out.

    I’m also reading the The Art of Game Design by Jesse Schell right now, and one of the ideas in game design is tuning the game for good game balance (Flow, really) so that challenge, novelty, levels, pacing, etc. all work well to keep the player engaged. Not that it doesn’t get considered in learning environments (“Hmm, I need to add an activity to break this up a little.”) but I think that game developers have a more sophisticated vocabulary for how they address it. Will need to think more on that…

    • finiteattentionspan

      Howdy (and what a great justification for coming late to a blog post; I never thought of it that way before :)

      I agree about letting students fill the gap – one of those ‘learning abhors an unclosed loop’ kind of things? And yes, out-of-class reading must have an obvious point, or no-one will do it (which makes me think that our struggle to get students to read in advance of seminars is perhaps about our failure to explain why it’s important).

      That paper is wonderful, thank you!

      I’d love to read your thoughts on the game design book (blog post, perhaps?) … it sounds very valuable. Just in the middle of Raph Koster’s Theory of Fun and I suspect some of it might overlap.

      Thanks again!

  6. Just found this.


    Take-away is near the bottom when he talks about ‘Site Analytics’. By doing his kind of PBL/Project-based learning, and sticking it all up online, he’s got himself a wicked proxy for measuring attention.

    Just saying.

  7. Pingback: Ask, Don’t Tell: the power of questions « Finite Attention Span

  8. Nif

    I am fascinated with your post and the comments. I fully believe in what you describe and want to be the kind of teacher that causes students to read on their own prior to class, when it’s not for a grade. :) I teach biology to 10th graders, and I am wondering if what you describe is applicable to the university level only, or is it possible with 15-16 year olds who have been trained to listen in class, take notes, regurgitate on tests, and only do the work that is for (what they consider a decent amount of) points. Suggestions?
    Thanks for your post and ideas.

    • Hey,

      Firstly, my sincere apologies for being so late to reply; been neglecting these parts lately! Too much marking.

      I am convinced that this kind of teaching can work well with any age of student, with the caveat that younger students will have less attention and fewer cognitive resources for you to play with. But 15-/16-year-olds would be fine, I’d imagine. You know, as much as they ever are ;)

      Our students coming to university now are not so different from the students you describe. Since it is expected in the UK that up to 50% of kids will go on to attend university, those who stay in the system are often trained (implicitly) to be the passive recipients of knowledge that you hint at in your own comment. So I say go wild!

      Because 16-year-olds are harder to wrangle than 18- to 20-year-olds (all those hormonal, peer-group social issues!) I guess you would need to be more alert to potential wrinkles and maybe have a more watertight plan (don’t be afraid to mix up the groups if you feel that your students are engaging in ‘social loafing’; put them with kids they don’t know or like so well and all those dynamics change instantly). But I also think that in some ways, teenagers are easier to engage than our occasionally ennui-laden undergraduate body — for example, turn it into a competition, one group against the others. Who can get most questions right? Or promise that the group who does best can hold a small quiz of their own, based on the material, towards the end of the class. Students seem to like being given the chance to be in charge.

      Is any of this helpful? Happy to keep talking — shout if you want to take it to email or anything.

      Best, and hang in there for the teaching you want to deliver!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s