Callusing blog header photo
Callusing's c-blog
Posts 0Blogs 7Following 0Followers 2



McMillen's Masterful Tutor


Even if everybody on the ice wore the exact same jerseys, any hockey fan could pick out Sidney Crosby in an instant. NHL players are obviously all a talented lot--that's why they get paid to do something millions of Canadians do for fun. But even by this standard, Crosby's movements seem unusually effortless. He skates with his head up, scanning all across the ice for open teammates, gaps between defenders, and shooting lanes. His rarely seems to focus on the puck, and yet he regularly performs mind-blowing feats of agility that few can match. If we didn't know better, it would seem like magic, or at the very least, some untouchable innate talent that made him a “born hockey player.”

Now if we watch a bit of this match from a recent Black Ops 2 World Championship, it doesn't take long to see this same sort of “magic.” While aiming and shooting is just about the only thing an average player ever thinks about, these guys make it look trivial--threats are spotted when so much as a leg sticks out, crosshairs snap instantly to their targets, and a few quick bursts almost always does the job. They seem to know where their opponents are before they see them, and over and over again, they pull off kills that us mere mortals could only do with an immense dose of luck. To do it this consistently is incomprehensible.

Science has given us an explanation for the skill of both Crosby and the Call of Duty champs. One of the modern understandings of talent, popularized by the likes of Malcolm Gladwell and Daniel Coyle, argues that the reason these experts make things look so effortless is because they are, for them, effortless. These actions are governed by what psychologists would call implicit thought, or what laypeople would call, instinct. To put it simply, Crosby has put so many thousands of hours of practice into puck control that his brain can handle these actions on autopilot. A Black Ops player has aimed at so many thousands of foes that it has become entirely automatic. And the result is that, with their brain freed from these menial tasks, these experts can focus on observation, judgment, and analysis, even when in situations that would overwhelm those of lesser skill.

We see this ease, this ability to perform at a level that is literally unbelievable, in nearly any complex skill, from the soccer player who can juggle a ball in ways that defy physics to the pianist who can play a flurry of 32nd notes with poise and subtlety. In almost every case, the road to gain this skill is long and arduous; in most cases, it's physically and mentally painful. But I believe that doesn't have to be the case. I believe that a tool has been developed that can teach high-level skill quickly, effectively, and painlessly. In fact, I hope to demonstrate why an indie game from two ex-Newgrounds developers is one of the best teachers of skill I've ever encountered.

The Meat of the Argument

Saying that Super Meat Boy requires precise input is like saying that Ronnie Coleman is a large man. It's technically true, but it's so much of an understatement that it completely misses the point. Super Meat Boy is a game of millisecond timing and millimeter inputs. Hold jump for two hundredths of a second too long, and your poor meat cube's contents are spewed across the level. Swivel the analog stick a couple degrees too far to the right or left, and your fate is sealed. To pass even the simplest levels requires robotic accuracy, and the more complex tests would be impossible without practice.

And yet progress is constant. The first attempt at an obstacle is often laughably bad. And the next attempt may not be much better. But eventually, you get a bit closer to success. And closer. And closer. Until, in a glorious moment of triumph, you succeed.

This first success is when the real magic starts to happen. Whether it took a dozen tries, fifty tries, or three hundred tries, the obstacle seems just a little bit easier after it's been beaten. And each time you pass it, it gets even easier. And easier. Until eventually, it becomes effortless. This obstacle that caused so much grief, this jump with error bounds that would make NASA break into a sweat, just sort of happens. Your brain and fingers go on autopilot, and performing the movement perfectly becomes as trivial as tying your shoes.

You can watch this happen on an SMB replay after success on a particularly tough level. The first runs at a difficult obstacle, whether a giant sawblade or a tiny platform, spawn a huge cloud of red squares that throw themselves at it from every direction. In the middle of this cloud, however, is a thin stream of runs that flow right around the danger. This stream represents effortless perfection, dozens of successful runs, each nearly identical to the one that preceded them. You couldn't be this precise if you tried--in fact, this can only happen because you no longer need to try. Your brain has been wired to do the work for you.

The parallels to the high-level skills mentioned above are obvious. A series of inputs or movements that seemed almost impossible at first gradually, with practice, becomes possible, then easy, then entirely effortless. By the end, a skill has been developed, and repeating it requires nary an ounce of effort. But there's a big difference between SMB and process required to gain most other skills--time. For a hockey player to perfect a wrist shot takes years, and for a guitarist to learn to even fret a single note consistently can take hours. But SMB drives a drastic evolution of skill in just a few minutes. Something incredible is happening.

Skill-building systems

Super Meat Boy is perfectly constructed to teach a complex skill. This comes down to 5 key characteristics:

 - Rules A standardized and unchanging set of rules that allows the player to learn and adapt

 - Feedback Direct, instructive, and clear feedback on not just if the player failed, but why

 - Repetition A design allowing dozens of repetitions in the span of a few minutes

 - Challenge A design that demands the most out of its players, while keeping success within reach

 - Flow Through a combination of the above, a style of gameplay that keeps players intrinsically motivated to keep playing

If any of these are missing, the game becomes a much less effective learning tool. It is only the presence of all five of these in one game that generates its impressive effect.


Super Meat Boy levels are dynamic. While the first world might have a few levels with nothing more than stationary platforms and sawblades, by the time you reach the fourth and fifth areas, you'll be weaving between balls of lava, dancing circles around monsters and rockets, and squeezing past slow-moving slugs. But despite all this movement, the way things move never changes. Meat Boy always runs the same speed and jumps the same height; a platform always crumbles after the same amount of time; a rocket always has the same turning radius. Even more importantly, the levels are, by design, not randomized...every single run starts with the entire level in an identical state.

The result of all this is that the world is governed by a strict set of rules that foster learning and adaptation. Take the screenshot below from a level named “Flood” in the fifth world. This was taken an instant after the start of the level, and I can already tell you which of these replays survived and which ones failed. That slug on the right starts at the same place every time, and slowly creeps to the right side of the screen. He's been placed precisely so that if you run as fast as possible to the right and jump as far as possible the instant the level starts, you can make it past him. The slowpokes in the back of the chain here are probably not going to make it.

This is, of course, just a single obstacle...learning to get past this slug doesn't say much about my “skill” in the game. But over time, it's easy to pick up on consistent properties of these obstacles that can be exploited. I can learn that, when I see a rocket launcher, I can climb a wall for a bit then bounce off to dodge it; I can learn that I've got long enough to do a small hop on a dissolving platform before I find myself without anything to stand on. I can learn how quickly a slug moves with enough accuracy that I can confidently judge in an instant where it will be half a second later.

If these obstacles were variable, progress would be possible, but much slower. Anybody who player soccer growing up probably remembers the difficult transition between playing with a size 3 soccer ball and a size 5...shooting a ball is still shooting a ball, but the complex muscle memory that had been honed through dozens or hundreds of hours of practice didn't quite work in this new environment. This change in the environment caused a temporary regression, and forces players to re-learn some of the skills they thought they had already mastered. It's not an impossible hurdle to overcome, but it does slow things down.

By standardizing the way the player interacts with the game and the obstacles they face throughout their playtime, SMB ensures that every minor skill learned has an application that stretches beyond a single obstacle on a single level.


The real world sucks at feedback. We can usually tell whether we've succeeded or failed, but the information is binary, a simple “yes” or “no”, with no insight into why things turned out the way they did. More importantly, there is a dearth of information about what could be done next time to change the outcome. When I drive a golf ball, I can gain a little information about what I did from the way the ball flies. But even this feedback system, which is more nuanced than most, does little more than narrow the cause down to a set of possible explanations. If, for example, I hit a steep slice, it could be because my hips weren't square, I hit the ball with the toe of my club, the shaft of my club is too flexible, my wrists aren't snapping enough, or that I put too much spin on the ball. If all I have to go on is the way my ball flies, it's impossible for me to figure out how to straighten out my next drive. While I can try to fix one of these many options on the next attempt, it's difficult to change just one aspect of my swing without unintentionally changing many others. The result is that even if I hit my next shot perfectly, it will be hard to tell why that shot went well and the previous shot went poorly. This makes progressive learning difficult.

In Super Meat Boy, feedback is precise and easy to understand, which gives it direct instructive power. Take the jump shown in the screenshot below, also from “Flood.” At least 19 attempts are shown here, and I'd estimate about half of them make it. But each one of those failures occurs for a very obvious reason. The guy circled in yellow, for example, will bounce off the wall he's on and end up too high on the left-hand wall to make it over the grub pile. This was an early attempt, and upon seeing this, I learned to skip the right wall entirely and go straight to the wall on the left. The one who just died in the screenshot tried to cut the jump short--a couple failures later, I'd managed to gauge exactly how long “jump” had to be held down to clear the pile. And so it goes, with each failed run meeting its demise due to a glaring and obvious player error. The game is saying more than “you failed”; it tells the player, “go harder on the jump button” or “take it a bit slower next time.” So next time I take it a bit slower or jump a little further, and get instant feedback on that as well...I either see myself get closer to passing the obstacle, or I don't, and I know I'm either improving in the right ways or I'm not.

Super Meat Boy is helped in this regard by its two-dimensional simplicity, which makes it easy to pick out precise causes and respond with constructive changes. But fancy golf swing coaches have managed to apply this feedback principle to a more complex sport as well. And the Australian olympic team famously designed complex feedback systems to help train their athletes before the 2000 Summer Games in Sydney, and walked away with a truckload of medals as a result. Feedback is essential for progress; with it, each failure is a step toward the next success, and without it, searching for success is like playing darts blindfolded--you might hit a bullseye, but it'd be hard to argue it was anything more than luck that caused it.


If you're trying to gain skill at something, there's two ways to make progress. One way is to use brute force--in other words, tee up 300 golf balls, start hacking away at them, and keep hitting until it “feels right”. Alternately, you can do what Coyle and Gladwell call “intensive practice.” Intensive practice trades speed and instinct for thought and contemplation; you perform a small movement, as small as a single dribble of a soccer ball, reflect on what went wrong or right, and then apply this knowledge to the next attempt. Sometimes this process happens naturally, like the way a Brazilian fusbal player slowly learns to feel the ball on their feet and understand the way it moves. Other times it's more formal; UFC super-coach Greg Jackson has become known for using precise movement analysis to optimize every aspect of his fighters' performance.

In intensive practice, each repetition is an opportunity to learn. More repetitions, then, means more opportunities for skill growth.

Let's say you want to spend a day at the driving range, and start with an ambitious bucket of 100 golf balls. In theory, you could go through these balls in less than 15 minutes if you were really trying. But assuming you want to practice intensively, and intend to take some time to reflect on each shot and improve your technique, it'll be difficult to go through more than two balls per minute. You'd be hard-pressed to finish your bucket in under an hour with any semblance of thought or control.

How about shooting a soccer ball? If you're lucky enough to bring ten balls to the field, and have a generous goalie who will feed shots back to you, you can get in a focused shot every ten seconds for a little while. But sometime before the 100th shot, you're going to start to tire, and your pace will slow. If you're dedicated, you'll get done within a half hour.

Compare this to Super Meat Boy. A single attempt rarely takes more than ten seconds, and both the reasons for failure and the means to improve are instantly recognizable. More importantly, there's no “reset” time whatsoever...less than a second after a failure, the next attempt can begin, without a need to set your feet or place a ball on a tee. Every instant in the game world is spent doing one of two things--attempting to pass a level or reflecting on what needs to be done to reach the goal. The result is that it's perfectly feasible to get in 5, 6, or even more attempts in the span of a minute, each deliberate and instructive. With no fatigue to speak of, 100 cracks at an SMB level can be done in ten minutes.

More practice means more opportunities to grow. More opportunities to grow will, over time, lead to faster growth. By presenting an environment that assaults the player with both opportunities to try their skill and feedback on those attempts, SMB fosters a growth potential that few others can touch.

Challenge and progress

While we enjoy the end products of learning, and like knowing a skill that we can show off to others or that improves our lives, the actual process of learning sucks. To learn a skill, we're forced to fail dozens or hundreds of times at the most menial and uninteresting tasks with no certainty of progress. Skill-building is not like putting experience points on a level bar, where each action and each attempt brings you visibly closer to a greater level of skill. It's more like following a trail through the woods with an bad compass...you know you're doing something, and you think you might be getting closer to what you're looking for, but there's no way to be sure. 

This makes learning very difficult and very disheartening, especially for those without external motivators to force them to stick with it. I vividly remember spending summer afternoons bouncing a soccer ball off my knee and watching it fly away from me over and over again. Eventually I just gave up, convinced I would never learn to juggle. I still haven't.

Super Meat Boy has motivation built right into its design. Progress, however menial, is very easy to see. Even on the toughest obstacles, slight improvements in technique bring you inches closer to success, constantly feeding a desire to improve while emphasizing that the challenge can be bested. And bigger goals are never far away--you always know a good run could have you 30 seconds away from complete victory. Since the goals never seems out of reach, it's easy to keep working through dozens of failures--after all, the next attempt could be the one that does it. As an exception that proves the rule, some of the few times SMB becomes disheartening are on the longer levels, where a usually manageable challenge suddenly seems much more difficult to overcome.

So while most practice in the real world is forced, and treated as a means to an end, SMB uses constant rewards and careful use of “carrots and sticks” to bring means and ends closer together.


Few academic theories have struck me like Mihaly Csikszentmihalyi's (Me-hall-ee Tsik-sent-me-high) concept of flow. Very roughly, it can be summed up in one sentence: The human brain is wired to love working at its limits. Most people, when asked about their greatest experiences, cite times they forgot about the outside world and were pushed to the limits of their skill set, yet felt in charge of whether they succeeded or failed. Surfers love the way the waves transport them to an elevated mental state, where everything irrelevant disappears, and every action becomes a precise and deliberate part of the battle to stay upright. Surgeons are addicted their jobs because it demands the most of them every day, and brings them into an environment where their actions have a direct and unassailable effect on their future.

Super Meat Boy is an exquisite producer of flow. It starts with the level of challenge. An easier game would be, by definition, less mentally demanding. While the upside is that you might be able to play it with beer-in-hand, the downside is that it would no longer generate the sense of stress and growth our brains relish. A much harder game, on the other hand, would do nothing more than frustrate. But the slowly-building challenge and simple core mechanics ensure that only rarely is a level much harder or easier than what the player is ready for. The skills almost always match the challenge.

The other part of flow, a sense of control over the outcome, comes from the well-established rules and lack of randomness described above. It's very obvious that success or failure in SMB is entirely the result of player skill, with nothing else to blame or credit for the way things turn out.

So Super Meat Boy sucks you in and demands that you forget about everything else while you play. It then constantly reminds you that skill alone will determine whether you succeed or fail. When you fail, you're driven to improve, since success always seems tantalizingly close. And when you succeed, you can celebrate a job well done, the significance of which is only magnified by the narrowed focus. This is flow incarnate.

Flow, in other words, is the glue that binds everything together. Your brain becomes engrossed in the game and is continuously supplied with interesting challenges and visible rewards. Every failure brings you closer to these rewards, and the rapid pace of progress keeps you in the action and focused. With all these pieces together, SMB is a deliriously enticing and powerful concoction...if any one disappeared, the entire system would fall apart.


I've used up about ten minutes of your time going through why I believe that Super Meat Boy is educational perfection. I've hopefully convinced you that this characteristic is nothing more than a few simple traits with a significance beyond the sum of their parts. A few of you may have thought this was pretty cool. More of you are probably wondering what the point is.

The skill that Super Meat Boy teaches is simple--it's nothing more than the ability to move a tiny little red square with legs and eyes around a series of obstacles. It's certainly a skill; while it's learned through mastery of specific levels, these any processes build a comprehensive, instinctual understanding of the game mechanics that can be applied to levels new and old. But with 4 buttons and a 2D world, the complexity of other video games puts it to shame, let alone the nuance of hitting a golf ball or swinging a tennis racket.

But the reason Super Meat Boy succeeds as a teacher is not because it is simple--it's because the skill-building tools it offers can match the complexity of the system that is being learned. In other words, despite its simplicity, if you take away any of the critical elements described above, Super Meat Boy fails.

What this means is that successful skill teaching depends not on simplicity, but on a learning environment that matches the demands of the system. In any realm, from a racing game to parkour to painting, utilization of these five key tools, in a way that's consistent with the environment, will lead to faster skill growth and greater enjoyment of the skill-building process. SMB is important not because of the fact that it teaches a skill well; it is important because it shows us how to teach skill well, and gives us a basic blueprint that we can apply to any other skill imaginable.

To wrap things up, I'll point to one example where ignorance of these principles backfired. Ninja Gaiden 1 and 2 would sit right next to Super Meat Boy on the pantheon of flow-inducing games, with an incredibly demanding, but highly predictable and rules-based combat system that always made it obvious why you screwed up. When Ninja Gaiden 3 came out last year, the rave reviews turned sour. On paper, nothing had changed...combat was still fast and fluid, generic enemies still came in droves, and Ryu still bounced around like a hypercaffeinated flea. But the devil here was in the details. The difficulty was lower, Ninpo magic was boosted to room-clearing potency, quick-time executions replaced much of the standard swordfighting, and the complexity of Ryu's abilities was slashed. The result was a loss of challenge, a loss of the incentive to build skill, and a loss of that critical feeling of player control; the game was allegedly so simple that rooms could be cleared without even looking at the screen. The first two games had demanded and rewarded a high level of player skill, and as a result cultivated the growth and generation of experts. The third title ignored the key characteristics that allowed this to happen, and failed miserably as a result. Not every game needs to be skill-based to succeed, but when it's the lifeblood of a series, developers would be wise to pay attention to this example.
Login to vote this up!


taterchimp   1
Arkhon   1
ShadeOfLight   1
Ben Davis   1
sirdavies   1



Please login (or) make a quick account (free)
to view and post comments.

 Login with Twitter

 Login with Dtoid

Three day old threads are only visible to verified humans - this helps our small community management team stay on top of spam

Sorry for the extra step!


About Callusingone of us since 11:20 AM on 06.07.2013

Recent grad, professional researcher, aspiring writer.