Okay, let’s do one more article on creativity and games, based on this question: Is it better to hand hold new players through a game tutorial to teach them all the mechanics and intricacies of a game, or is it better to let them figure things out on their own?
The “tutorial level” has become so ubiquitous in video game design that it seems really odd when a game does not go to to painful lengths to make sure you get a slow, measured introduction to every single game mechanic. This is presumably so you don’t get confused about what the Y button does and burst into tears. For example, I started playing the game FTL (http://www.ftlgame.com/) earlier this week and while the game does offer a brief totorial and many tooltips, it expects a fair amount from you in terms of learning how to play the game on your own. My first half hour with the game consisted mainly of a steady stream of expletives and mutterings like “Why would I ever spend money on door upgrades?” and “Wait, why are all these rooms turning pink?” and “OH GOD! WHY IS THAT ON FIRE? WHAT FIRE? HOW FIRE? …WHAT DO YOU MEAN GAME OVER?”
FTL (or “Faster Than Light” for the cool kids) gives you a brief overview, then tosses you to the space mantis/slug/rock men and expects you to figure the rest out yourself.
Eventually, though, I got into the groove and realized that for a game like FTL, part of the experience should be experimenting with new things, paying attention, and learning how to maximize your chances of survival on your own. It’s not dissimilar to systems driven, sandbox games like Minecraft or Terraria in that way: they just dump you into a system and tell you that figuring it out is half the fun. (The other half is feeling superior to people who complain about it not being spoon fed to them.)
This all reminded me about another psychology experiment I learned about from Jonah Lehrer’s recent book, Imagine: How Creativity Works. In a 2011 paper impressively entitled “The Double-Edged Sword of Pedagogy: Instruction Limits Spontaneous Exploration and Discovery” Elizabeth Bonawitz and her colleagues set out to examine how different modes of instruction affect how creative people get in their exploration of a new system. And by “people” I mean “toddlers.” Yes, toddlers are people; I looked it up. And also by “system” I mean “toy.” Work with me here.
The researchers invited kids visiting a science museum to check out a new toy, except not in that creepy way that you hear about on prime time news shows. The toy was a crazy homemade contraption consisting of tubes that did different things like squeaking, lighting up, and playing music. It’s important that these functions were not obvious and required some experimentation to discover. For some children, the experimenter took out the toy and said something like “Woah, look my badass new toy! Check it out!” Then she yanked on a tube to demonstrate how to make it squeak and finished up with “See that? This is how my toy works!”
For other children, the experimenter took out the toy, acted like she was seeing it for the first time, then pretended to accidentally make it squeak. She then feigned surprise (children are very gullible, it turns out) and said something like “OMGWTF? Did you see that? Let me try to do that!” then made it squeak again. For kids in all conditions, the experimenter gave the toy to the kid and finished by saying “Wow, isn’t that cool? I’m going to let you play and see if you can figure out how the toy works.”
Picture of the toy, taken from Bonawitz et al. (2011).
So, the key points here are that the toy did multiple things, but only one thing (the squeaking) was revealed. For some kids it was explicitly demonstrated and for others it was serendipitously discovered.
What the researchers found was that relative to those in other conditions, children who were given instructions on how to make the toy squeak played with it for shorter amounts of time, did fewer unique actions with it, and discovered fewer of the toy’s other functions.
Now, I understand that most of you reading this are not toddlers, but I think this has clear implications for video games. Because when we are given a thing and told “here is how it works” that presentation tends to constrain the list of things that we consider doing with it. We explore less and are less creative. Our brains tend to take the paths of least resistance, and heavy handed demonstrations create a nice easy rut for our thoughts to follow.
It’s Minecraft. Figure out what you want to do.
Sometimes this is great, as with simple games designed around mastery of a few skills. But for games dependent on the interaction of multiple systems, options, strategies, or approaches, detailed tutorials may hurt the player and their long-term experience with the game. Booting up a game like Minecraft, blinking a few times, and then saying “Okay, what happens if I do …this?” is a great experience and facilitating that approach is central to the appeal of the game. Like the kids who were told “this is a squeaky toy, here’s how to make it squeak,” players who get their hands held through an hour of tutorials are being mentally primed to consider only what they’re shown. Accident, serendipity, and an occasional bit of rudderless flailing about are sometimes necessary for creativity and exploration.
REFERENCES
Lehrer, J. (2012). Imagine: How Creativity Works. Boston, MA: Houghton Mifflin Harcourt.
Bonawitz, E., Shafto, P. Gweon, H. Goodman, N., Spelke, E. & Schulz, L. (2011). The Double-Edged Sword of Pedagogy: Instruction Limits Spontaneous Exploration and Discovery. Cognition 120, 422-430.
Your bit about FTL is so very true. Initially, I was shocked at how brief the tutorial is. But now, I *think* as a result, every little thing I discover makes me appreciate the discovery.
Contrary to my expectations of frustration, I keep returning to the damnedable thing. Just One More Game… and all that. I still haven’t won a game on the Easy difficulty, but I’m sure that I can. It’s just going to take more time and fumbling.
Yeah, I’m pretty much still in the same boat, er, ship. It’s a good game to take quick breaks with.
“Your bit about FTL is so very true.”
Except…not really ? “why would i ever upgrade doors” – because there is a tooltip that explicitly says it helps with fires and boarding crews ? And just by reading that you might, just might, surmise that there are such things as fires and boarding in the game ? So its not exactly the fault of tutorial, but a common enough reading failure.
It is one thing when the game actually tells you nothing about some aspect of itself, and something quite different when a player just outright refuses to read in-game text and then whines “stupid game, why did this happen, where do i go, what do i do now ?”
I criticized The MDA Framework a while back, one of my main gripes being the hand-wavy bits about “fun” being a component of games, how to categorize it, etc.
Cutting right to the chase, I’ve been thinking a lot recently about why that rubbed me the wrong way and have been toying around with the idea that games are mediums of discovery, wherein “fun” crudely sums up the feeling of the process of discovering systems and narrative.
Systems discovery involves finding out the core interaction points with the game (mechanics in MDA), i.e. “what can I do?”, finding out how the systems work together to form something greater than their sum (dynamics in MDA or colloquially “gameplay”), as well as a number of other elements, such as system optimization and skill improvement. Narrative discovery includes explicit narrative (the story), implicit narrative (the emergent player-driven story), as well as reflexive elements of self-discovery and meta elements of sharing, discussion, and community.
FTL instructs the user so far as in the interaction points, but leaves it up to the player to discover when to do what and how best to optimize. It is no surprise that creativity is often defined as a way to use two seemingly unrelated things in a novel way. The high of experimentation and discovery of how underlying subsystems interact is the essence of the ever elusive “fun” that games are unique in delivering.
FTL does a great job of pacing that systems discovery, both along the course of a single game session, and, true to the roguelike tradition, across multiple brutal deaths. This discovery is at the core of the evolving player narrative– “I didn’t know I could get killed by THAT!”, once again, one of the central tenets of roguelikes/roguelike-likes.
Great post! I think there are various reasons why we play games, and the system optimization one is definitely one of them. It often gets lost amongst other stuff like narrative, competition, etc.
Pingback: The Alexandrian » Blog Archive » Thought of the Day – Introductory Adventures
Pingback: Links for September 29, 2012 | Andrzej's Links
Pingback: Superlicious | Superlevel
Discovery is great! But…
..women tent to dislike at least some form of discovery. Women rather like to know how things work upfront, before diving in. I’m a firm believer of discovery as one of the main pillars of fun. That said, the “tutorial” given to the toddlers is in both cases the same, but only uses different “wording”. Now that’s interesting and is something I’ll definitely keep in mind.
I would disagree, mostly because I am a woman and love to learn on my own, through trial and error discovery. I would be interested to hear/see the cited evidence of why women don’t like discovery learning.
A better gauge would be childhood experiences and toys played with growing up. Imagine two kids are given some Legos. One kid gets the step by step instructions on how to build this castle. The other kid only gets an image of the final castle without the steps to get there. One requires more discovery and trial and error learning. Depending on what type of child you were, and how you preferred to learn, this may predict the type of game genres that are of interest.
I can’t cite this immediately, but through my work experience I’ve seen this plenty of times. Of course, the problem with generalizing is that it isn’t applicable to everyone. I work in the casual game space and through usability test I’ve conducted women tend to want to know how a game works (controls, goal, basic rules) before diving in. Men, on the other hand want to discover this step by themselves too. This doesn’t mean that women don’t like discovery, or even the particular discovery of certain game elements, but it seems they feel more comfortable if they at least know something. Why this is, I don’t know. That said, I’m a firm believer of discovery being one of the reason why we all play (men or women)
Incidentally, I am acquainted with this interesting book by Donald Norman, “The Design of Everyday Things” through one of my grad classes.
An interesting book about people’s conceptual model of everyday things, how they form them through naive physics and the design pitfalls of new technologies frustrating people because it’s not easily discernible.
It is perhaps implementing tutorials were meant to avoid the frustration of bad design. Norman did touch on something that indicates FTL is a great game and everyday thing, at least it did not taught us helplessness. At least, you are still trying and getting timely feedback, see visible things that does what it does and nothing else.
I keep hearing about that book. I should really pick it up.
Great article if a little bit brief!
Absolutely agree that we should encourage exploration in the player’s discovery of game mechanics. The caveat of course is that the games top layer of mechanics needs to be robust and simple enough to understand well before the player gets into the nuanced aspects of the game.
What I found out making an iOS game is that when playtesting openly that the tutorial was the make or break deal for many players, either the tutorial was too hard and testers put down the game after a couple attempts or they managed to get through and began to enjoy the game.
Originally, I wanted the tutorials to be very open in nature, just letting the player funnel themselves through the level and protect them enough so that they can discover the mechanics themselves. What testing showed though was that many people would be be hesitant to face the challenge even though it was clear that was the only way through or make wild assumptions about how the game was meant to work. Also many players would misinterpret the results of what happened.
What I was eventually forced to do was do instruct the player to specifically “Go here and watch what happens.” followed by “Well done! This is what just happened!”.
Knowing that helped us to refine the tutorial until we were getting around a 7 out of 10 randomly selected people passing through the tutorials without help. Not really what I wanted as it’s contrary to my personal learning style but it was effective for this particular design of game. I think it’s also because we were putting people into an environment where they weren’t getting into the experience by choice, i mean clearly when you download or buy a game there is a drive for you to make your investment take you places, that certainly helps exploratory learning.
Granted, I also think it’s a product of how the industry markets games in general. Lowest common denominator tactics invariably result in some measure of mediocrisy in the products that paymasters seem to believe is an acceptable loss to generate greater sales.
I think the example in the article of fire doors pretty much sums up the need for informative tutorials.
A game commits suicide when it either tricks you into failing or lets you fail and then says “You should have done X”
Some people like a bit of help, some need a lot of help, others like to find everything out for themselves, but I suspect you’ll struggle to find anyone who enjoys the experience of failing after 20 hours of gameplay when something they weren’t told about kills them.
“Oh yeah, that boss will kill you AND wipe your quick and auto saves. Should have told you that earlier, sorry dude”
I couldn’t disagree with you more there. A theory that has been hanging around game design like a bad smell for a few years now is that there is a perfect game balance where the player constantly feels challenged but never fails. What tosh!
There is nothing wrong with losing a game. Ideally, a game should be fun whether you win or you lose. Unfortunately, a generation of games developed with the idea in mind that losing is some sort of punishment and should be avoided at all costs by the devs has bred an attitude in gamers that mirrors it.
We’re seeing the rise of a new way of thinking about game design at last which involves not hand-holding players through a game but allowing them the creativity to discover, through losing, through experience and (hopefully) through having fun. FTL’s a damn fine example – it’s tons of fun when you lose, it’s tons of fun when you win. Better, when you reach the stage of winning the game, that doesn’t mean the end of the road because you’re still rather likely to face just as much of a challenge on the next playthrough.
I’m not against losing, I suck much more than I rock and if losing were a problem I’d be locked up.
It’s ‘how’ you lose rather than the fact of losing itself. I’d rather understand how I just got killed (failed a dodge, chose the wrong tunnel, forgot to cast ‘resist acid’) than just be handed a ‘Game Over’ screen without explanation. There are a number of ways to give that explanation and tutorials are only 1 of them. Checking every single option by trial and error without information is probably the worst way.
Anecdote – Tomb raider 2 (I think) had a jump you couldn’t make, literally. Lara just slid off the edge of the block without grabbing. This wasn’t even a death though, you dropped into water and could climb up and fail again, on what looked very much like a bug. Then, randomly trying every option no matter how stupid and illogical, I found that a dive instead of a normal jump worked. Lara made the jump without sliding off the forcefield.
That issue never came up again
Least Satisying Solution Ever.
I know you’re not advocating blindfolded trial and error in games, but a tutorial has it’s uses. Fortunately even if the companies decide to stop using them you can usually find a video or 2 on youtube to find out why the hell Blob Monsters can take you out every time (knives vs blobs, why the hell would I even try that? Oh, it works. Fine, whatever)
Pingback: The Sunday Papers | Rock, Paper, Shotgun
I feel like the problem is more one of design, hand-holding, and instruction rather than just exploration. Well-made a game should provide instructions, not tutorials, that explain how everything works. If I can perform an action it should be listed and I should be told how to do it and what it does. In other words if I cast “Fireball” it shoots out a fireball to range X which costs Y mana and does Z damage. This should not be hidden information where I have to puzzle out what button to push or how much the spell costs. That isn’t helping matter or some sort of fun exploration, it’s just shoddy documentation.
What is fun, however, is how your known abilities interact with one another and the environment. Perhaps my fireball can ignite an oil slick I can also cast or my freeze spell will freeze a lake solid letting me walk across. This is encouraging and interesting to work with. I know my own capabilities and limitations, but I can explore the world and test how I can interact with it.
Then again, I’m the type who, even as a child, never played a game without first reading the manual. Even on arcade games I would dutifully read the little card to learn how to play while friends would rush to mash on buttons and then gape “Hey, how’d you do that?” “Well, it’s right here in the instructions.” I’d reply. When I buy furniture from IKEA I read the directions twice, catalog the parts, and then refer to each step as I perform it. I take notes and precise measurements when I cook rather than just doing whatever I “feel”. But then I’m a scientist and precision, reproducibility, and rules are key. Not to mention exploration.
Minecraft is a ridiculously addicting game, when you have ideas for what to do. However, when there is no inspirations or cool ideas, it becomes a fairly pointless game. But, for some reason when I get a great idea for something to build, I’m right back into it again.
I agree. Minecraft is a great example of a game with a blank canvas, and being able to just go forth and explore without any constraints. I do think that if there weren’t any tutorials available, there may be more limited players though. As it would easily get quite frustrating trying to figure out how things work in the beginning.
Pingback: On Making a Tutorial | Tankeflod
Tutorials are killing my enthusiasm for new games. At the very least, make them optionable. Games for me are about discovery, the a-ha moments, the pride of figuring things out using nothing but your intellect. In this time and age with game faqs, meplays, walkthroughs easely available and now with the newer systems even integrated social help options why oh why is it nescessary to so blatantly piss on the joy of exploration with a mandatory hour long hand-holding….Really, if games of today could hold my dick while I urinate they would kill themselves over the job.
Looking outside of games, perhaps a small-house movie should have a five min introduction too, every classical or contemporary piece of music should begin with a message from the composer, every splash painting should be covered in a curtain stating what you really should think you are looking at. My God. Let games be as any other art form about exploration and the unique interpretation of each individual who experience it. An introduction that stifles all creativity is not a great beginning.
Pingback: 游戏教程是否会扼杀玩家的创造力？ - 爪游控
Pingback: Exploring Tutorials in Video Games
Pingback: Exploring Tutorials in Video Games | My Blog
Pingback: Superlicious | Superlevel
Pingback: FEZ: Teaching gameplay through exploration and a lot of instructions | Next Level
Pingback: Stephen Calender Programming Blog