Deep Learning, Deliberate Practice, and Desirable Difficulties: An Interview with Patti Shank


Deep Learning, Deliberate Practice, and Desirable Difficulties Image

In this article, we’re going to investigate how to help workers develop job skills through deep learning, deliberate practice, and evidence-based training methods known as desirable difficulties.

Sounds pretty exciting, no?

Well hold on to your hat, friend, because it’s more exciting than just that.

We’re going to do this by talking with Dr. Patti Shank, one of the most informed and generous learning researchers out there, and someone from whom we’ve learned a lot about learning over time. We’ll say it below as well, but many thanks to Patti for sharing her time and knowledge with us.

You’ve got two options for taking this all in. Read the article or scroll to the bottom and watch a recorded video of our conversation.

Facilitating Deep Learning with Deliberate Practice and Other Desirable Difficulties

Convergence Training: Hi, everybody, and welcome again. This is Jeff Dalto with Convergence Training and we’re back with another installment in our audiocast/podcast/webcast/interview series.

We have a really exciting guest today and I’m really excited. We’re in the world again and we have Dr. Patti Shank from Leaning Peaks with us, and if you don’t know Dr. Patti Shank or haven’t heard the name, she’s somebody I learned about and learned a ton from going way back to a book she wrote about how to use one of the Articulate elearning tools, and I follow her on social media and learn a lot there. So I’m really excited and with that I’d like to say hi to Patti.

Patti, how are you? Welcome!

Dr. Patti Shank: Hi, Jeff. I’m doing really well. It’s sunny after a few days of snow here in Denver, CO, and as I was telling you earlier, any day without sun is a day that people in Colorado are antsy and anxious. So I’m having a good day and I’m glad to be here with you.

Convergence Training: Glad to hear it. And there’s a metaphorical aspect to that as well, because I know you just got over a cold, so I’m glad you’re feeling better and thanks for being here.

Dr. Patti Shank: Thank you. And thanks for waiting, because I really did sound awful. For a few days, I didn’t sound like anything, because I just croaked.

A Little about Dr. Patti Shank

Convergence Training: Well, you’re sounding fine now, so I’m glad.

So I wonder if, before we get into our question list, if you could tell the people who will be listening or reading a little bit about yourself and what you do.

Dr. Patti Shank: I think one of the most important things to know about me is that I’ve spent most of my life actually DOING training and development, building it, working with companies to build it. But I started as a corporate trainer–well, I started a longer time ago as something else, as most of us did–but I’ve been a corporate trainer for a long time and ran a corporate training and health education department on the east coast.

So if you hear an east coast accent, and I say I’m from Denver, that’s why.

So that’s my background, and probably everything I do is focused on pragmatic ways of improving the outcomes of what we do. And that comes from doing it. I’m not someone who has just read things. I think through “Can we do that, can we actually do those things?” So the last four or five years, I decided to read the research on a variety of training topics–and I don’t mean just learning, I mean TRAINING topics–and figure out what the research says will give us our best outcomes. And I had no idea–zero idea–if anyone would read this. I had been told by a previous employer that what I wanted to do was just not going to fly; nobody was going to be interested in it. And I did it anyway.

So I spent the last four years writing three books, I’ve got two more possibly in the works, and maybe even another one, but I’ve got someone who’s actually going to help me find funding, because those three years were completely unpaid. And you get paid from selling books, but at ten dollars a book it’s hard to fund the writing.

Side note: Friends, run–don’t walk–and buy one of these books now: Write and Organize for Deeper Learning; Practice and Feedback for Deeper Learning; and Manage Memory for Deeper Learning.

So that’s what I’m doing right now, and I’m also working with organizations who have really important training outcomes that they need to meet. Like I’ll give you an example. I was working with energy engineers and they had a problem that HAD to be solved. It was a skills-based problem, and I worked with them for about four or five months to figure out what are the methods that would best get them the outcomes they needed. And I just LOVE those projects.

Should Learning Be Fun?

Convergence Training: That sounds good.

So, two quick things: One, I’m surprised to hear that you got that negative initial feedback on your new turn of career, and hats off to you for continuing anyway. I’m really interested in those books and I’ll ask more about them later. And two, just so you know, we had some little intermittent audio drop-outs, so when I’m typing this transcript I’ll get back to you for any missing audio.

OK, so you agreed to come in today and talk about some things I’ve seen you writing about and talking about in a recorded conference presentation, and it basically gets down to, in big-picture terms, how do people learn and do they learn from fun or do they learn from something else? And I wonder if you could talk about the role of fun training and whether we should try to make learning fun or what we should be focusing our efforts on?

Dr. Patti Shank: Yeah. It’s a really good question, and I actually did a debate in Berlin with people who are well-known on this issue. And I decided to take the side of no. It had to be yes or no. And the way they phrased it was, should all learning be fun?

I took no, and someone else took yes.

And here’s the thing. Learning CAN be fun. It can be loads of fun. But here’s what we know. We know that learning needs to be effective, and effective learning, effective instruction, especially instruction that’s to get us to a certain desired outcome, is often not fun.

And that doesn’t mean there’s anything wrong with it. It’s just that the research shows that once we get past certain easier-to-understand concepts, the amount of effort required is high and often difficult.

And so–a lot of this comes from expertise research–and the expertise research shows us that learning can be grueling. And to get to any level of expertise is often grueling. And the people who get to that level of expertise are those who are willing to put in the time and effort despite the fact that it’s not fun.

Here’s a good example that came out of the Berlin talk: Olympic athletes, skaters, speed skaters, and figure skaters. What they have to go through in order to get to that level of expertise is not that much fun. But once you get to that level of expertise, it’s lots of fun.

For instance, I spend a lot of time reading research and taking notes. And when I first started doing that, it was not the least bit fun. And I had to push myself, because it’s written in ways that are obtuse and difficult. And after doing that for 3-4 years, I can now read something and it’s not painful anymore.

So learning does not need to be fun in order to be effective. And effective learning often isn’t fun. So it’s not that learning shouldn’t be fun, but when we need people to get to a certain level of skill, it may not be fun and they’ve got to get there anyway.

And you and I have to design so that they get there. And we have to use the best methods. And some of those methods are not that much fun.

For instance, people talk about gamifying instruction to make all learning fun, but unfortunately gamification is good for some learning but less good for other learning.

Convergence Training: So, fun.

It’s great that you brought up gamification, which as you’re saying might be appropriate in some cases and not in other cases.

And I guess since gamification is in some senses–I don’t mean to disparage it–but it has a buzzword aspect to it…

Dr. Patti Shank: It does.

Convergence Training: …that brings to mind maybe THE buzzword, which is engaging as well. And I wonder if in this context if you could talk about engaging, and when engaging and when it’s not and even what it means?

Dr. Patti Shank: Yeah. Here’s the thing with the word engagement. It’s a buzzword that’s been given a meaning that it doesn’t mean.

So engagement in the learning and work literature talks about the level of effort people are willing to put in. So engagement equals effort. Now, there are parts of engagement that have to do with the level of interest. But it’s not the only thing. And interest and engagement don’t necessarily mean the same thing.

Convergence Training: Great, great. And often times, that even gets watered down to engagement equals fun.

Dr. Patti Shank: Right, exactly. And so fun may not work. Engagement IS necessary for learning, in that effort is necessary for learning. But effort can be put forth even without interest, although it is easier if there is interest.

So that’s why some of the research talks about making sure there’s value in what you’re doing, what you’re training, for the person who is being trained. They have to see the value in what they’re doing.

Those things have more to do with engagement than just making it fun. And I’ll give you an example. My ex-husband is an electrical engineer. And engineers don’t want fun–or their definition of fun is not the same as yours and mine. So he once came home and told me about some training he had taken, some safety training actually. And it was a cross-section of a building. And you had to find all the things that were not safe and click on them, and you had two minutes to find those things. Well, he thought that was ridiculous. So what he did, was he did a grid search. He went up and down the screen with his mouse, like this, and then across, until he found everything. And that was it. He was like “I only have two minutes, I don’t see the point, everything is really tiny, I really can’t see what’s on that desk. So I know I have to find 17 things before two minutes.” So he just did a grid search, because that’s what an engineer would do.

So what’s fun to one person and what’s fun to another person? I mean, I know the person who built that had to have spent a lot of time. But the important thing here was that we wanted people to know what’s not safe in what they were seeing. And was that the best way to do that? It’s a good question, but you need to know your audience. You need to know what THEY find valuable, and he did not find that valuable.

What Is Deep Learning and How to Facilitate It

Convergence Training: So I like three things there.

One, there’s probably a funny joke to be made about ex-husbands there.

Two, I love the emphasis on and the idea of hazard identification training, which I think is tricky.

And then there was getting to know your audience.

So that’s a good lead-in to my next question. I’ve seen you talking about superficial learning and what you discussed earlier with your mention of expertise–I think you’re calling it deep learning.

Could you tell us a little bit about what superficial learning is and what deep learning is?

Dr. Patti Shank: Right.

And these days, when we use the words deep learning, we have to be careful because machine learning also talks about deep learning, and I’m not talking about the exact same thing.

So superficial learning is when you learn something for the purpose of being able to parrot it back on a test. Or when you can recite facts and define things. And it’s not bad. It’s required for us to learn facts and concepts and all these things, somewhat out of context at times, so that we can go to deeper learning, which is learning for application. And you’re absolutely right, it does tie into expertise. We have to have a base knowledge of facts and concepts and definitions before we can learn the next step and learn for application. So those are the definitions I’m using for those, and they come from research.

And my position is we need to train for application. And training for application starts with a more shallow level of learning, with definitions, facts, concepts, and things like that, that you’ll need in order to understand things, but it simply doesn’t go far enough. And if that’s all we’re doing, then we’re wasting resources, because people can’t apply with just that.

What Are Desirable Difficulties?

Convergence Training: Great.

You know I like that notion of the importance of training people on that superficial stuff–facts and concepts–but then teaching people to apply them. And although the spotlight is on you right now, I read a really good book by Dr. Ruth Colvin Clark called Developing Technical Training that addresses that really well.

Side note: Read more about different types of learning (and corresponding different types of training) here.

So to get people past that superficial learning, those facts and concepts, to help them with application on the job or wherever, and to help them with the development of deep learning and expertise that we’re talking about, one of the things we can read about here is something called desirable difficulties. I wonder if you can talk to us about that?

Side note: In addition to the discussion and linked materials below, you can also read more about desirable difficulties here.

Dr. Patti Shank: Great. That’s exactly what I was talking about earlier.

Dr. Bjork, who wrote about desirable difficulties, tells us that there are certain things in remembering, understanding, and application that are not fun, that are difficult, and they help us remember and they help us use what we learned. And if we don’t add that into our training, people are going to be less likely to understand and less likely to use and apply.

Convergence Training: Good. We have a blog article on desirable difficulties, and for people who are reading this interview, I’ll add a link to it (there it was), but could you give us an idea what some of these desirable difficulties are, including the fact that some of them are perhaps a little counter-intuitive or unexpected?

Dr. Patti Shank: Yeah. There are a couple of big ones.

One is stating what you understand about what you just learned in your own words. And the other one is having to recall–adding activities into the training that force you to recall what you just learned. And I do that in a lot of my online training–there’s a lot of recall activities. And people sometimes are like “Well, you could have covered more if you didn’t do that,” and here’s Bjork’s point. Covering isn’t what we need to do. What we need to do is have people remember and then be able to apply it. We can’t apply without remembering.

Convergence Training: So, great points. That brings to mind three things.

One is that there’s a difference between covering and helping people retain and apply information.

Two, I like that you talk about recall during training, and that’s something that you can have them do after the training as well in the form of refresher training.

And third, I wonder if you call talk about the difference between recall and something a little simpler, like recognition.

Side note: In addition to the discussion below, you may find this earlier discussion I had with Dr. Will Thalheimer on the spaced learning, recall, and recognition interesting as well.

Dr. Patti Shank: Right.

So I’m actually writing a manual right now for multiple-choice questions, so that brought that up for me.

There are two different types of multiple-choice questions. In one, you have to input the answer. And in the other, you get to select from a bunch of choices.

And recognizing is simpler than being able to actually supply an answer.

So multiple choice often is written–it doesn’t HAVE to be written, but it’s often written–in the form of “I recognize which of these is the correct answer.” It’s not really very deep.

But multiple choice questions CAN be written in the form where you have to be able to understand in order to correctly answer the question.

Convergence Training: Gotcha.

So that superficial multiple choice question where you’re recognizing is not as desirably difficult as one that would require you to recall.

Dr. Patti Shank: Even better, to understand or apply.

I’ll give you an example, one I came up with is you see a chart showing how many children are in families in a neighborhood. And then you have to figure out the mean number of kids in the neighborhood. That’s an application. And you’re not able to just recognize, you’re not going to be able to pick the right answer without being able to apply.

And so that takes us back to writing learning objectives that are on the same level as what people need to be able to do on the job. And that’s hard. That’s the basis for us picking activities. If we want people to be able to do something on the job, then what are the activities?

People think we’re here to build content. We’re not. We’re here to get people to meet the learning objectives. And so selection of learning objectives is really critical. Most of the time it’s not define, and it’s not list: it’s to be able to categorize, or make decisions about, or select the most appropriate form of action.

Convergence Training: Right.

I guess going back to that learning objective thing, the simplest and most unhelpful one would be something like know. And so to side-step know, they end up with list or define…

Dr. Patti Shank: Right. So they don’t have to use understand or know. And here’s another one–appreciate.

Convergence Training: Appreciate, right.

But nobody actually goes to work to list, define, or appreciate.

Dr. Patti Shank: They do not. They usually have to choose things, and analyze things, and decide which to use–like, for learning content, we have to decide which learning content to include and which to exclude. There’s no list objective there.

So hard stuff.

Deliberate Practice

Convergence Training: Right.

So we have this interesting idea of desirable difficulties, which kind of forces our brain to grind a little harder but they help with retention and later transfer onto the job in learning.

In addition, another concept alongside desirable difficulties and often discussed in the development of expertise, which is important for the job, is deliberate practice.

I wonder if you could tell us a little bit about that.

Side note: in addition to the discussion below, you can read more about deliberate practice here.

Dr. Patti Shank: Right.

So I talk about practice and feedback in my practice and feedback book, but deliberate practice goes even further.

Deliberate practice is needed to get you to the next place you need to go. So I have this level of expertise, and I need to get a little bit further. And it’s very goal-centered, and it selects activities (generally with feedback from a mentor who has more skill and is able to give that feedback) on the things you’re doing wrong and practicing over again. And that’s another area where learning is hard.

This happens, let’s say, in medical school all the time. When people are learning to be surgeons, they’re going to have people right there all the time asking questions, and “What are you going to do next?” and “Wait a minute, that’s not appropriate, how about thinking a little more deeply,” and just activities that specifically get people to the next outcome. And they’re planned and you do that until you reach that next outcome, and then you pick a next one.

This is used in music and sports all the time. Music and sports have natural “next” things that you need to accomplish. And that’s true of all skills. Just working through the difficulties. Let’s say you’re doing a double axle, and you’ve only done a single axle. and so your mentor is going to work with you: what is the thing that you’re doing wrong? And you keep doing it over and over until you get there.

Convergence Training: Right.

I’ve taken saxophone lessons and had to go through really tedious exercises that clearly were designed to make it easier to move my fingers in a certain way. And I’m taking online lessons from Fender right now on guitar, and I’m doing the same thing.

And this idea of deliberate practice is often associated with a guy named Anders Ericsson, and in his book he’s talking about the player Ray Allen, who’s a great shooter. And people would often say “Oh, he was born with this amazing skill,” and Allen mentions “I get upset about that, because I worked really hard and hard to do all that boring stuff, worrying about where my elbow was over and over and over again while my coach was watching and commenting,” so yes, all that stuff–the grind.

You pointed out, I guess, three key points: (1) you have got a goal, (2) you work really hard on it on specific little tiny things, (3) and you’re getting some kind of feedback, either just by seeing the consequences of the action or ideally by a more experienced observer who can tell you what you’re doing right and wrong.

Dr. Patti Shank: Right.

And I didn’t talk about that in the practice and feedback book because I don’t even think most training folks realize how important just practice is on what they’re being taught. And most of the training I’ve reviewed has FAR too little practice and far too much content.

So the type of practice you and I are talking about, deliberate practice, is an order of magnitude higher even. To fix the little stuff–you’re talking about where your fingers are and where your elbows are. Skiing is a great example of that. Where your edges are on either ski, where they’re pointed, where your feet are pointed, how your knees are, how much over the skiis you are or how far back. You know, it’s just those little things that keep you from getting to the next level.

And I’d like to see us get to the point where we put in more practice, because content is not as important as practice.

Convergence Training: Right.

And I think sometimes because these examples come easily, playing saxophone or shooting a basketball, it’s easy forget that these same things apply to things we do at work. Like you can have a novice trainer write a whole bunch of learning objectives and get a whole bunch of feedback on that. Or operate that machine a whole bunch of times and learn about that over time as well.

Dr. Patti Shank: Right.

No, I think it applies to EVERYTHING we do. But too many of us don’t realize that.

That’s our actual job, to get people to a certain outcome so they can get past that with further training.

Wrapping It Up, Other Learning Researchers, Additional Resources, Connecting with Patti

Convergence Training: OK, great.

So we talked about fun and engaging, which was helpful.

We talked about superficial and deep learning, and you’re sticking up for superficial learning while saying it has its place but we need to do more.

We talked about desirable difficulties and deliberate practice.

Anything else that you would add on this basic topic?

Dr. Patti Shank: The thing that occurs to me is that people in our field don’t realize that we’re not here to build content. We’re here to build skill and knowledge.

Content is one of the things that helps with that. It’s necessary but it’s completely insufficient.

I guess my main point is our primary purpose here is to give people the knowledge and skills that they need for the job, and we actually need a real mind shift around what our job is.

Convergence Training: Great, great.

So, to that point, for people who’d like to learn more about this topic, could you tell us more about books by you that people could read, where they can see you at upcoming conferences, and how they can follow you on social media?

Dr. Patti Shank: The books are…I’ve written three of them.

The first one came about because it’s a huge problem I saw in trainers and training writers in general. And that is problems in writing with the clarity needed for instruction. And so the first book is Write and Organize for Deeper Learning.

The second book is Practice and Feedback for Deeper Learning. Another problem I see trainers having issues with.

And the last one came out of questions from the first two, and it’s Managing Memory for Deeper Learning.

The ones I have on the back burner–I’ve got a couple. One is about what the research says about media for deeper learning. And Karen Hyder, a very close colleague of mine, would like me to write with her one on synchronous training for virtual classrooms. And I’ve asked a friend of mine who’s an expert on task analysis to do a short, concise guide on that, because the average trainer is not going to do a full task analysis, but what are the most important pieces so we understand the job. Because you cannot write good training, you cannot build good training and practice activities without understanding the job. So those are on my list.

Ruth Clark’s books are really good, Will Thalheimer, Clark Quinn has a new book out about some myths about training. But my stuff was written to be completely practical. And I’m not saying the others haven’t done that as well. But that was an established rule when I was writing: these all have to be practical and easy to apply.

Convergence Training: And that IS one hallmark I’ve noticed of your stuff–it’s practical and easy to apply and accessible.

Dr. Patti Shank: Right, and if it’s not practical, or if it’s very hard to apply, even if it’s important, I tend not to talk about it, because it won’t get used.

Convergence Training: Gotcha.

So I like what you talked about the task analysis earlier, saying “Hey, our job is not to build content but it’s to build skills that people will apply on the job,” and you can’t really do that unless you know what skills people need to apply.

Dr. Patti Shank: Right. Exactly.

Convergence Training: Alright. I encourage people to go out and get your first three books, and I look forward to the next three as well. I know that one you’re working on about media, I’ve seen some of the things you’ve written for that, talking about how video is effective for certain things, including teaching procedures. Is that right?

Dr. Patti Shank: Right. And it’s not good for everything.

One of the most well-known, and it may not be well-known to the audience, but if you are describing or explaining a diagram, or you’re just showing something that’s on the screen and you’re pointing out things, maybe with call-outs or whatever, it is much better to explain complex pictures with audio. And the research is really clear–use audio to explain complex images. Because, here’s the problem: if you explain it in text, the person reading has to go from the text to the diagram, back and forth. And I realize that for people who have hearing loss, that’s not going to be helpful, but that’s a central finding of dual-coding research that’s mentioned all the time: explain in audio.

Convergence Training: Which I guess goes back to Mayer’s Multimedia Principles and, as you mentioned, this idea that we have dual pr

Want to Know More?

Reach out and a Vector Solutions representative will respond back to help answer any questions you might have.