Complex learning, step by step
March 25th, 2009
Series: Ten Steps to Complex Learning
I’ve been (slowly) reading Ten Steps to Complex Learning, by Jeroen J. G. van Merriënboer and Paul A. Kirschner. The subtitle explains why: A Systematic Approach to Four-Component Instructional Design.
I read a lot about the death of instructional design, the end of training, and the New Jerusalem of learning that’s due any day. Certainly a lot of superstition and nonsense gets daubed with the label “instructional design,” like a kind of cognitive Clearasil. Still, I can’t help think that few people are going to learn to manage power-generation stations, conduct clinical trials, sell aircraft engines, or produce FEMA-acceptable flood elevation certificates solely through self-guided learning.
So I decided to plow through this book, which I’ve described with a bit of humor as being written in a language very much like English: the prose is dense, and very academic. So far it’s worth the effort, and I’m going to summarize key parts here.
(Key part: something I pay enough attention to that I make a note on paper as I’m reading. This is an ancient custom among my people.)
Van Merriënboer and Kirschner aren’t shy:
The fundamental problem facing the field of instructional design these days is the inabiliity of education and training to achieve transfer of learning.
Which is something like AIG not being able to actually insure anything, isn’t it?
One point the authors make is that most complex skills require the learner to coordinate from a range of “qualitatively different constituent skills.” That last phrase is important to them: not only is the whole of a complex task more than its parts, but the constituent skills are not parts of the larger task but aspects of it. They’re not sub-skills, which you add together to make up the Big Skill.
Which, they argue, makes the analytic approach of many traditional instructional design approaches counter-productive. For example, what they call the transfer paradox comes into play: the instructional methods that work best for isolated objectives often work poorly for integrated objectives.
To make that plainer: we spend too much time fiddling around with nice, clear, low-level objectives. Then we lack time and money (and, perhaps, the will) to develop integrated learning. Then we wonder why the training/learning function has such a dismal reputation.
But those isolated ones are what we tend to grab onto, because it’s easier to design around them, easier to create test items, and easier to cram them into an LMS (“Lessons Mean Simplicity”).
Learning to use the Amtrak reservation system is a complicated task, but maybe not all that complex. Learning to act on traveler’s questions is also complicated. Developing training for either set of skills is inherently less difficult than developing holistic training for an effective Amtrak reservation agent–but that’s what Amtrak’s really looking for.
The usual answer to the problem often seems to be “watchful waiting.” The performers go from training to the job, and we hope that their random encounters with reality end up filling the gaps.
Van Merriënboer and Kirschner want to grapple directly with such complex learning problems. The model they advocate sees four main components to a learning blueprint:
- The learning tasks that someone needs to master. (Strictly speaking, I’d say these are the on-the-job tasks which the person currently doesn’t know how to do, but it’s not my model.)
- The supportive information that comes into play when you’re working with skills that are performed differently from problem to problem. These skills, which they call schema-based, benefit from things like mental models of the overall domain (e.g., pharma research) and cognitive strategies for working in that domain.
- The procedural information that guides those skills that are performed the same way from problem to problem. This is the how-to knowledge (e.g., using the clinical trials database) that’s a routine part of the overall task.
- Part-task practice to strengthen and automate certain “recurrent constituent skills.”
Van Merriënboer and Kirschner argue that people can only perform certain constituent skills (which are aspects of the larger task, remember) if those people have a certain level of knowledge about the larger domain. “Select an appropriate database,” as they point out, doesn’t make any sense if you don’t know what makes databases appropriate to the search you’d like to perform.
To foster integration and avoid compartmentalization, their model includes an emphasis on inductive learning: you as the learner work with specific problems so you build and improve mental models for the principles behind those specific problems.
…all learning tasks [should] differ from each other on all dimensions that also differ in the real world, such as the context…in which the task is performed, the way in which the task is presented, the saliency of the defining characteristics, and so forth. This allows the learners to abstract more general infromation from the details of each single task.
That’s how we learn a great deal of what we know. And, yes, a good deal of that happens informally, though I don’t see that as an argument for not trying to create learning situations when the informal can happen more predictably and more rapidly.
Related to this idea, the authors advocate always having learners work with whole tasks. That might mean starting with simple cases or examples. Other approaches include providing support (say, a process overview for the clinical-trial system) and guidance (a job aid for forming database queries). They also make use of task classes, by which they mean categories of tasks. In their ongoing database-query example, one task class has to do with performing searches when the concepts are clear, the keywords are in a specific database, when the search involves few terms, and where the result includes only a limited number of articles.
I’d call that the “clear, simple search” class.
You can imagine the other extreme: a poorly phrased request involving unclear concepts, with little knowledge of the appropriate databases, calling for complex search queries and producing large numbers of relevant articles which require further analysis.
How many task classes do you need? That seems to depend on the range of variation between the Clear Simple Search class and the Nightmare Search class.
There’s a lot more going on; without intending to, I guess I’m starting another series.
The posts in this series:
- Complex learning, step by step (that's this post)
- Complex learning (coffee on the side)
- Ten little steps, and how One grew
- Problem solving, scaffolding, and varied practice
- Step 2: sequencing tasks, or, what next?
- Clusters, chains, and part-task sequencing
- Step 3: performance objectives (the how of the what)
- Criteria for objectives–also, values and attitudes
- Step 4: supportive info (by design)
- Learning to learn (an elaboration)
- Step 5: cognitive strategies (when you don’t know what to do)
- Step 6: (thinking about) mental models
- Step 7: procedural info, or, how to handle routine
- Procedural in practice
- Step 8: cognitive rules, or, when there IS a right way
- Step 9: prequisites, or, ya gotta start somewhere
- Step 10: part-task practice (getting better at getting faster)
- You? Auto? Practice.
- Media’s role in complex learning
- Self-directed learning: stepping out on your own
- Where do the Ten Steps lead?
March 25th, 2009 at 7:51 am
“We hope that their random encounters with reality end up filling the gaps.”
This may be where we need to consider creating richer contexts. More of a case-study approach using actual cases.
For instance – instead of just training the reservation system – put the reservation system in the context of an actual encounter. Incorporate customer service questions + the nuts and bolts of using the application.
So much application training is “here’s how you click the buttons.” But there are processes and resources and people things surrounding the use of the tool.
Providing that context makes the tool make more sense during the training – and may help fill in some of the gap.
Though we will never fully replace the effectiveness of reality.
March 25th, 2009 at 8:22 am
Wendy, we’re on the same page. One point van Merriënboer and Kirschner make (I may start calling them vM & K) is that a lot of instructional design has tended to break down complex tasks into a cluster of subtasks, then focus on the subtasks. One thing that gets lost is context.
Your reservation examples are good, and they could incorporate even the essential practice elements for rule-based skills. For example, passengers often ask about the schedule from point A to point B. One way to vary the input (help strengthen the schema) is to phrase the request in different ways: “How long does it take?…I’d like to go… Is there a train?… What time can I leave for…?” One way to build the skill of reading the system display (a rule-based skill) is to include origin/destination pairs that involve one or more connections, or more than one train on a route without connections.
(“How long does it take to go from Boston to Seattle?” involves at least two connections; “Is there a train from Philadelphia to New York that will get me there by 2?” involves several possible trains without any connection.)
Another point–one I want to come back to in a future post–is that the instructional strategies for rule-based skills are different from those for schema-based skills. So the immersive environment by itself isn’t going to address those rule-based skills well; you’ve got to make that happen It’s like learning a foreign language: you need to spend time using it in rich contexts (talking with native speakers, reading in that language, watching movies); you also need to study grammar, because it’s a lot harder to derive the forms of verbs just from context.
March 25th, 2009 at 8:26 am
Dave – great conversation starter! I struggle with this type of training for complex management tasks in our retail environment every day. I am looking forward to a great series from you on this topic.
I agree with Wendy that perhaps we should be focusing our energy on designing the range of classes from Clear Simple Search to Nightmare Search. I have been leaning toward short, powerful context-based situational experiences for the learners. Many of these have been role-based, but we are not restricted to such role plays. My thought at this time is to put the supportive and procedural information as links to books, articles, videos, etc. that explain the management theory and practice and to focus my efforts on designing the “classes” probably starting with the Clear Simple Search first.
Any feedback on this approach would be helpful! Thanks again for starting a GREAT conversation!
John “Zed” Zurovchak
March 25th, 2009 at 8:33 am
John: I’d read an earlier, syntactically dense paper by vM & B; I had to make notes as I went along, because I kept getting lost. This book is a significant improvement, though it wouldn’t be everyone’s idea of a beach read. I hope I can do it justice.
I’m still not sure I understand how constituent skills are “aspects” of the larger task rather than “subtasks,” except for the central idea that the subtasks on their own don’t simply add up to the larger task. I was stuck on task classes, too, until I understood the Clear Simple Search and Nightmare Search as the extremes of a spectrum. vM&K seem to imply that you look at that spectrum and make some judgments about what would be intermediate points of sufficient importance to justify another task class (and thus more exercises, activities, etc.).
It also seems clear to me, though I haven’t said so, that I don’t see all this stuff happening in formal, classroom-style environments. I wouldn’t rule those out, mind you, but here on earth you’ll have to consider costs and benefits. vM & K view the Ten Steps as an instructional design model that you’d typically use “to develop training programs of substantial duration…several weeks to several years.”
“Several weeks” is enough to make some people blanch–but you’re not going to have the basic ability to use Amtrak’s reservation system without 20 – 30 hours of hands-on instruction plus additional, on-the-job practice. When you combine those essentially rule-based computer interactions with knowledge of routes, customer service skills, accounting requirements, and so on, then, yes, you do have several weeks. People just aren’t spending those weeks in a classroom setting, and the challenge to instructional designers (and to management) is to provide people with the best tools and the best learning opportunities so they can acquire the integrated skills as efficiently as possible.
March 25th, 2009 at 8:37 am
Awesome stuff Dave. Two thoughts immediately came to mind:
John Carroll — so much of this reminds me of his research on software tasks: moving from the specific to the general, having a variety of cases, and helping students see the forest for the trees by giving them the big picture.
Siebel Call Center Training — in a former life, I was an expert in Siebel and PeopleSoft and consequently built a lot CBT, WBT, simulations, and the like to support these sort of initiatives. One time I was doing a consulting gig for a large company and was asked to review their existing training which wasn’t working.
The basic gist was that they were giving folks all sorts of time with systems training, and a decent chunk of time with phone skills and conversation trees and all that, but (and it’s a big ‘ol but), they never put them together. So the first time that the new hire was actually trying to use the system *while* having a conversation was live with a customer. Talk about trial by fire.
What they wanted were compassionate, efficient, and knowledgeable call center reps. What they developed were call center reps who could use software well *or* have a cordial and fluid conversation. When employees tried to marry these skills on the job, it was a mess.
This seems like it goes to the heart of the argument you raise about “integrated learning.” They trained the pieces parts perfectly. They had nice clean objectives, but in the end no one could do the work, because they assumed integration of these skills would happen on their own (magic? fairy dust?). Once they started training integration skills and giving people time to practice these skills, overall performance shot up. Shocking, I know.
Needless to say, I think the book you are reading is spot on. Thanks for sharing and for the Cliff Notes version.
March 25th, 2009 at 8:53 am
Dave, you could probably start a very successful consultancy with pixie dust, as long as you called it something very contemporary like “Open Source Learning 2.0.”
I have no doubt there is something of “art” in all complex skills, by which I mean that some people seem to integrate the elements more readily. But that’s a lot like Philip Henslowe (the historical character played by Geoffrey Rush in Shakespeare in Love) saying, “It all comes together. Nobody knows how.” Henslowe didn’t mean you could throw just any gang of people together and come up with Romeo and Juliet — or even Ethel, the Pirate King’s Daughter.
This could be a long haul — I’ve read about 1/10 of the book, and had to prune my comments or no one would read them.
March 25th, 2009 at 1:11 pm
Dave, you are providing a valuable service to ISD here. I started reading about 4CID many years ago and found it interesting but was never able to integrate it into my work. You’d think that vM&K would be able to find a better editor, in order to get their point across.
March 25th, 2009 at 3:08 pm
Looking forward to the other 9/10s, Dave. Like Harold, I struggled with 4CID, and ultimately chose to look at patterns design and language (Christopher Alexander etc.) instead as a way to cope with complex learning. I’m glad you are offering a chance to revisit.
March 25th, 2009 at 4:36 pm
Dave, great stuff. I’ve not read the book, but a paper and a presentation, and I cite Van Merrienboer in my book. (And reading academic prose builds character, as if you aren’t enough of one.)
I agree it’s not for designing the formal classroom, in fact I think it argues strongly for games, er, immersive learning simulations (hence, the cite). My simplified version of their work is my focus on ‘decisions’ that people need to be able to make, not the knowledge. And that bit you quote about changing contexts is very much a point I make (e.g. my recent Broken ID series), covering all corners of the transfer ’space’.
As with others, I greatly appreciate you doing this!
March 25th, 2009 at 9:36 pm
Harold: I don’t know either author, other than an article on 4C/ID that van M published in a 2002 issue of Educational Technology Research and Development (vol. 50, number 2). I’m finding Ten Steps both more thoroughly developed and clearer.
I had a lot of downtime today, waiting with a relative at various medical facilities, so finished the three chapters. This slow pace of my own is helping me follow the overall argument.
Shanta: the bulk of the book devotes one chapter to each of the ten steps, but there’s a useful organizing structure that will probably be my next post in this series. I found lots of stuff clicking into place. (For the 34 pages I’ve read, I have 13 pages of handwritten notes.)
Clark: you’re right about the decisions, not the knowledge, I think. One statement I noted verbatim:
“Learners should not invest all their cognitive resources in performing the task [meaning, while learning], but should also invest sufficient mental efforts in genuine learning, that is, schema construction and rule automation.”
Why? Because those things are required for transfer to occur. In other words, they’re arguing that for the complex tasks that the Ten Steps address, transfer to the job won’t happen if people don’t build mental models for the domain in question and don’t develop fluency with using rule-based skills.
To all who have commented: I’m genuinely surprised and a little nonplussed by your enthusiasm. Reading the book was something I chose to do; writing about it was one way for me to work at understanding it. I had no idea anyone would be half as interested as you have been.
March 26th, 2009 at 8:01 am
[...] Complex learning, step by step | Dave’s Whiteboard | Dave Ferguson | 25 March 2009 [...]
March 27th, 2009 at 10:45 am
[...] In his comment on the previous post in this series, Dave Wilkins talked about call center workers who could either use the software well, or connect well with customers, but not both–because their training never integrated the two. [...]
April 9th, 2009 at 5:57 pm
I can’t help relating your discussion of complex learning to what I did for much of my career. I taught writing to students who weren’t in college to learn to write. Some teachers, especially those who were poor writers themselves, wanted me to teach grammar – to students who had been thoroughly taught that they couldn’t write during their previous 12 years of schooling.
Instead, I tried to get the students to write something they cared about, did triage in figuring out what most interfered with their communicating by writing, and dealt with one or two most significant problems. Then I got them to share their writing with each other, because the worst reader is the person (usually a teacher)who already knows the content and is just looking for mistakes. Anyone learning to dance (or to write) in front of a critical audience is only performing, not communicating, and will, almost inevitable, stumble.
Even weak grammar and poor spelling rarely prevents messages from getting through, (though it offends some who care more about perfection than communication). Not knowing what an audience needs to know first really blocks communication.
You don’t get a much more complex learning task than writing, and piecemeal teaching only works with those who are already highly skilled.
April 9th, 2009 at 6:04 pm
Joan, this is a great example. Writing’s a very complex task, and the average adult has integrated it so well — overlearned at least the basics — that the nuances have completely disappeared.
A perfect example of the curse of knowledge.
You can be brought back to earth quickly by trying too learn to write in another language, where the structures and patterns of your first language almost certainly get in the way.
It seems to me that you focused on the whole task–writing a complete piece, whatever it was–and found simple instances of it to use.
Anybody who thinks grammar rules are obvious should try explaining when to use articles, and when not (“my son is in college; my cousin’s in the hospital”).
August 27th, 2009 at 1:15 pm
[...] before I ever heard of Ten Steps to Complex Learning, I developed my own principle for helping people in organizations learn how to use [...]
October 18th, 2009 at 8:59 pm
[...] decision case learning, anchored instruction and the whole task learning design methods found in Jeroen van Merrienboer’s 4C/ID work. Here are some links to design approaches that are based on real world learning [...]