Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 22, 2014
arrowPress Releases
October 22, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

What are rewards? Decoding the Game Ep.2
by Nils Pihl on 10/27/13 08:28:00 pm   Expert Blogs   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


In the second episode of Decoding the Game we explore what rewards really are, introducing new concepts and terminology that help us better understand game design and the pitfalls of gamification.

This is the second episode out of five, so make sure you visit and subscribe to

Also, you might enjoy Why Gamification is a Dirty Word.


Hello and welcome to the second episode of Decoding the game, where we discuss behavioral engineering, the interesting intersection of behavioral psychology, game theory and memetics. 

I'm your host, Nils, and I'm a behavioral engineer. This episode I'll try to answer the question: What are rewards? But first, I'd like to start by making a controversial claim, and this one is for you, Gabe Zichermann: Counter to popular belief, points are not rewards. Let me explain:

If play is voluntarily performing an action for a perceived reward, understanding what rewards really are is going to be important. In popular discourse, rewards are often thought of as something that the designer gives to the player. Maybe that is a result of how we use our language - we like to say that we GIVE someone a reward for good behavior. But this definition, and this kind of language, suffers from a serious pitfall - we end up having to explain why rewards are sometimes not rewarding.

If a point itself is a reward, and rewards are motivating and fun, we quickly run into issues. If we make a game consisting of a single button that awards you one million points every time you click it, and that's it, no leaderboards, no challenge - that game should still be fun, by that definition - so, why isn't it?

To explain why this is, I'd like to introduce some new terminology that will help us dissect and understand rewards: The concepts of Currency and Tokens. 

The currency of a reward is why you're engaged - it's that feeling of mastery, or belonging, competition or discovery that makes the game enjoyable to you.  It's probably the reason you decided to play the game in the first place. A token, on the other hand, is a quantifiable representation of that currency. 

So, if we're playing Tetris for the currency of feeling mastery, a point in Tetris would be a token of that currency - and so would a position on a leaderboard. A new friend on Facebook is a token of the currency "social belonging and connectedness" that made you engage in the first place. 

Points, badges and leaderboards are not rewards in and of themselves.

So, what is it we GIVE someone when they perform well? We give them awards, and that way we don't run into the logical paradox of rewards that are not rewarding. An award does not have to be rewarding. 
What this teaches us is that points or badges or achievements will only feel rewarding if they represent a currency that we value. This is why gamification fails more often than it succeeds, and why Gabe Zichermann is wrong when he says "the presence of key game mechanics, such as points, badges, levels, challenges, leaderboards, rewards, and onboarding, are signals that a game is taking place."

The metric for measuring your progress in a game is not what makes the game. Implementing a leaderboard on your website will probably not make people buy your product. A badge is no substitute for quality and substance. 

But we can do more with these new terms, because it turns out that currencies can have different attributes. For one, a currency can either be finite or infinite. To explain what finite and infinite currencies are, I invite you to remember that old classic: the Pinball machine. You'll remember that Pinball machines had a fixed number of digits that it could use to display your score, meaning that there was a theoretical, and sometimes attainable, maximum score. For some people, reaching that maximum score was a highly motivating and enjoyable experience - but for others, it was a frustrating and involuntary end to their game. Some players are more likely to engage with finite currencies, while others prefer infinite currencies, and a player's preference can be highly predictive of their behavior.

If you're still not clear on what the difference is between a finite and an infinite currency, I'd like you to picture the following. Imagine three people that are each engaged in playfully walking. The first person wants to see if she can walk all the way to Mount Everest, and each step along the way is a small token, a quantifiable representation, of her progress. Once she reaches Mount Everest, she's done, and there's no point taking another step - She's won the game and another step would not be rewarding. She's playing a finite currency game.

The second person has a very different mission - she wants to see how far she can walk. Again, each step, mile or hour is a quantifiable measurement of her progress - but it will always make sense for her to take another step. Another step would always be rewarding. She's playing an infinite currency game.

The third person wants to see if he can walk further than either of the other two. Intuitively, this might seem like an infinite currency game, because the second person is playing an infinite game - but it is actually a finite currency game. Once the third person has taken one step further than the second person, he's done, and taking another step would not be rewarding.

If you watched last week's episode you'll remember that rewards can either be intrinsic or instrumental, and this is also a crucial distinction for currencies. Different currencies engage different players, and engender different player behaviors, and monetize in different ways. The kind of currency that the player is pursuing can become one of the most predictive things about their behavior. Tetris players that aim for a high score often take more risks than players that are trying to keep the game going. 

In versions of Tetris that award you with more points for removing more lines at once, players that aim for a high score will often use a strategy dependent on getting the 4x1 piece into place, removing 4 lines at once. This is a risky strategy that is likely to end the game early, and is less likely to be pursued by players that aim to keep the game going. 

As such, the tokens you choose to display in your interface can select for different behaviors. You can predict that putting a timer, with a time leaderboard, in a Tetris game will yield measurably different play styles than a version that displays points.

So to summarize, a point is only rewarding when it is a token of a currency that the player values.

What currencies does your favorite game have? Are they finite or infinite, intrinsic or instrumental? Which ones did you prefer? Share your thoughts with us over at Gamasutra. The link is in the description.

Thank you for watching, and make sure you subscribe if you want to know more about behavioral psychology, game theory and memetics. My name is Nils, and I hope that you'll join us next time as we continue decoding the game.

Related Jobs

Nix Hydra
Nix Hydra — Los Angeles, California, United States

Art Director
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Programmer
University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer


Robin Lindh Nilsson
profile image
Very interesting and clear! Keep up with this great work :D

Curtiss Murphy
profile image
What is the reason for using the term 'instrumental' vice extrinsic? Most talks, papers, and research I've seen makes the distinction between intrinsic (inside ... of) and extrinsic (outside).

Nils Pihl
profile image
The intrinsic/extrinsic distinction tells you something about the mechanics of the reward - where is it from? In my opinion, that distinction doesn't aid you very well in making predictions (which is the point of behavioral engineering).

The intrinsic/instrumental distinction on the other hand, borrows from value theory philosophy, and tells you something about the VALUE of the reward.

Note: The word "intrinsic" doesn't mean the same thing in both cases.

Even though most research in the field of psychology uses the former distinction more often (which I believe is a missed opportunity/mistake), you are always free to reinterpret their experimental results through a different lens. The candle problem is still the candle problem, regardless of what terms they use in their own paper, and the data is data.

The intrinsic/instrumental distinction makes better sense of some common cognitive biases, makes better predictions, and is more meaningful to track in your analytics.

I'll try to do a long form follow up post on your question, because it is one I get a lot, and it is a good and meaningful question. Thank you for bringing it up! I'll write a proper response as soon as I can. Want me to notify you by email when it is done?

Darren Tomlyn
profile image
Oh dear - the OP didn't really start off all that well. It's obvious what you're aiming for by the end - but it needs to exist on a more solid foundation.

Most of the problems we have exist for a simple reason - we don't understand what information words, (such as game, competition etc.) truly represent in relation to each other (and the language as a whole). (The fact that some people don't realise the language even has rules governing what information it represents is also a large part of the problem.)

The information that's important to recognise and understand, in order to build the perception upon that the OP is aiming for, is represented by the following words:

Game, competition and play (which also means recognising work aswell).

The most important problem, which the OP is based upon, is an inconsistent definition of play, either as behaviour (used as a verb) or applied as the state of such behaviour (used as a noun). (E.g. I'm playing/it is play.)

All human behaviour is either productive or not. The problem with the OP's definition of play, is that it leaves no room whatsoever for non-productive behaviour. This is a problem, because that is exactly what we use, and therefore need, play itself to represent. We therefore use work to represent productive behaviour.

What is non-productive behaviour?

It's easier to begin with a description of productive behaviour - (work). If we do something because of what it produces, be it a reward, actual product or state, then it's what we call work. A lot of people behave in a particular manner for money, (a job), but that is only a part of what work itself represents - for any and all productive behaviour, is work.

But what if we behave in a particular manner, purely for the actual behaviour itself, regardless of whether or not it's productive?

If it's not productive, then why would we do that? Why would we behave in such a manner?

Because we find it enjoyable.

This is what play is - non-productive behaviour we take part in because it's enjoyable.

So, by trying to make something non-productive, productive - by making it so that it's not about the behaviour itself, but what it produces, (a reward), what you're talking about is turning play into work. This is a really big problem with some of the games being designed today, because of the lack of understanding of what they're actually doing.

Another symptom, is that people perceive competition by what is being competed for, too, rather than the act of actually competing for such a thing.

Now, one of the reasons why such a perception has happened, is that people do not understand and recognise that games exist independently of both work and play - of being productive or non-productive - and that (what we now call) games have always been used for either. They mistake the use of play to describe that act of taking part in such an activity (playing a game) for it being non-productive. But what we call games have ALWAYS been played (taken part in) for work (productive reasons) - e.g. training/selection purposes, making the pieces of information represented by the word play, in those ways, completely unrelated to each other.


Gamification, however, is a problem all to itself. It has a use and place, but only when representing something far more specific than a lot of people perceive it as.

Gamification as representing elements found in games is as inconsistent and as problematic as it gets, for language as a whole - (not just English).


Because it exists to label (an) independent cause/causes, as and by it's/their subjective effect.

Games are an activity/event/application of things that happen. EVERY type of element games can use to enable such an activity, can be used and therefore exist for other reasons, including other activities and behaviour.

ONLY when such elements are part of such an activity, or are part of, or form, a collection to enable such a specific activity to exist, are they labelled as games. To try and label any such elements as a game, or related to a game/games without involving the specific activity itself, is to make a mockery of the breadth and depth of human behaviour that can use such elements without needing this specific activity. (We don't label woodworking 'tableworking' for a good reason, even if that is part of what wood can be used for.)

The main things the elements used for games can also be used for, are:

Play, (non-productive behaviour) (including toys), puzzles and competitions.

Of course, one of the main reasons we have this particular problem, is that games are not fully recognised in relation to puzzles and competitions (or even play) in such a consistent manner.

And so we have problems with the term 'gamification', because there's nothing related to games (as an activity) that it makes sense to represent, and is therefore only causing confusion.

So what can and should it be used for?

As an application of 'game theory'. Which is also a problem - a misnomer - because it's also not truly about games, either.

What game theory is truly about, is the mathematical modelling and study of COMPETITIVE behaviour: even if it might have been intended to be purely about games, that's no longer what it is at all. Because of this, it has relevance in everything from psychology to biology and beyond - since life, by its very nature, involves competitive behaviour, (as do games, but they form only a small part of what game-theory covers).

It's with this in mind that gamification has to be recognised - that it's about competitive behaviour, rather than games - and is also, therefore, a misnomer.

What gamification truly means, is the use of competition to enable and/or promote other behaviour - (often through the use of rewards to be/being competed for).

We can see how this can be related to games, specifically, since they are also competitive activities, and even use similar elements, but it doesn't have to be related at all - since competition is far more fundamental and therefore applicable to far more than just games. Again, games can be affected by this, but are not part of its cause.

The main problem people seem to have with this definition, is recognising the difference between competing for a particular outcome/reward, and basic productive behaviour, itself (work) - though gamification involves work, (making some behaviour productive), it doesn't mean they're the same thing.

Nils Pihl
profile image
UPDATE: My sincere apologies to the Gamasutra community. As a relative newcomer here, I didn't know not to engage in this conversation.

Darren, you are misrepresenting my view to fit an earlier narrative. I recognize your arguments from reading your comments on other peoples' articles, but you seem to miss that you and I are actually quite often in agreement with each other - if you stop distorting what I say.

I've read a lot of your collective writing on Gamasutra, and I'm familiar with your argument. It is unfortunate that you have jumped to conclusions regarding my argument, and lumped me in with previous arguments you've had.

If the player is engaging in the activity in order to reach a certain outcome (greater than "having won", including some state change that will further the player's position outside the game's context), the activity is work. I think we agree about that, and you'd label the "outcome" as "productive". Where I say the outcome has instrumental value, you say that the outcome is a product. Am I right?

If the player is engaging in the activity because the activity is enjoyable, any potential outcomes (productivity) don't change that what is happening is play. Productivity can be an accidental byproduct of play. Correct?

We're saying largely the same thing. Correct me if I'm wrong.

That being said, I am not a fan of the productive/non-productive distinction. I still prefer my terminology to yours, because you have still not stipulated what "productive" means - which seems like it will end up being a value judgement - which is not terribly useful.

I'd like to deconstruct the first part of your comment at greater length, but since we're already talking past each other, maybe it would be better to have a discussion in person? If you're not available for that, let me know, and I'll find the time to write you a lengthier response.

Also, I'm no fan of gamification, but I think you misrepresent it by saying it is about competition. Certainly, many implementations of gamification do include competitive elements, but not all. It's a sweeping statement that doesn't do anything to further the discussion.

EDIT: Clarification, it seems we agree about a lot of things that I say, but I disagree about a lot of things you say. Nothing in this comment is meant to be construed in any way as any kind of support for your wall-of-text series of blog posts where you make outrageous claims like "It stops being a game if someone cheats."

Darren Tomlyn
profile image
"If the player is engaging in the activity because the activity is enjoyable, any potential outcomes (productivity) don't change that what is happening is play. Productivity can be an accidental byproduct of play. Correct?"

Is not quite what I'm saying. Enjoyment is WHY we play, not what play IS. Understanding this is a very big problem, currently, and is affecting the games that are being made and designed in a very fundamental way - the OP also suffers because of this.

It's all about perception - if you know what you're doing is productive, then it's work, regardless of whether or not you enjoy it. One of the main ways in which we get people to learn, however, is of course the opposite - convincing people that what they're doing is non-productive, even though it is. This can be a problem, however, if you do not fully understand how to make that happen - and rewards are not it, since your using a product to enable the behaviour itself - deliberately making it productive to enable such behaviour to happen.

Yes, it's possible for some behaviour to be productive even if it isn't perceived to be - but rarely the other way round, since we tend to recognise such gains and rewards rather well - but it's all a matter of subjective perception and application that is inherent to humanity itself and all human behaviour in general - just because I see something as play, doesn't mean you have to, aswell - only that our understanding of work and play is consistent - which is currently a big problem.

If you can get your head round to understanding how and why work and play - productive and non-productive behaviour - are related to, and describe ALL human behaviour, including games, aswell as competition and gamification, then you'll be in a far better place when it comes to describing what it is you see in regards to how and why different types of rewards are used to enable other types of behaviour, especially within games, and what the further ramifications of such rewards are in relation to games themselves.

This should all be obvious to understand, since work and play are opposites - if play was defined as something done that is/was enjoyable, then work could NEVER be enjoyable at all - which is completely inconsistent with how the words are used.

If you don't understand the word productive - talk to an economist. What productive relates to is work, and so it should be no surprise that matters dealing with work - the economy etc. - are where the main understanding and description of such terms can be found.

The difference between productive and non-productive behaviour is whether or not the behaviour happens by itself, (because it's enjoyable), or whether it produces something. The nature of what is produced does not matter. As I said, though, whether or not you perceive some behaviour as productive or not, is subjective. Unfortunately, because people are getting confused between being non-productive and enjoyable, they're calling enjoyable productive behaviour, play, when it's not - it's work.

What I'm giving as the definition of 'gamification' is what it was originally used for and as - that has since become corrupted by people confusing such elements with games themselves, (which isn't surprising since game theory itself suffers from such problems because of its label), even though they exist separately, as I said, and therefore causes a lot of problems - also related to games, puzzles, competitions, work and play etc.. Since such problems and inconsistencies exist because it's not how the language functions, it should not be allowed to continue, and will only create much larger problems in the future. (You could argue that a recognised difference between games, puzzles and competitions is already under threat, which has tangible effects already - (especially when related to gambling.))

Nils Pihl
profile image
You are much more condescending than your argument merits. Before you bother saying that I don't understand, please be aware that I understand your argument very well - the problem here is that you don't understand your own argument as well as you pretend. Don't bother saying I don't understand you, because you've proven that you haven't really read or listened to my argument. I've read your walls of text, and I know what you're saying, and what you're saying is patently absurd.

You start by contradicting yourself (from previous blog posts) and saying absolutely absurd things, which I will now demonstrate with a series of reductio ad absurdum:

"It's all about perception - if you know what you're doing is productive, then it's work, regardless of whether or not you enjoy it."

Playing League of Legends produces proprietary data that is used to increase Riot's revenue. You are now aware that playing is productive. League of Legends can now not be play, to you, by your definition. You can't pretend that you are unaware of this data being created, so it is unarguably a part of your perception that your actions within League of Legends are productive.

Ah, but it is not productive to YOU, you say? Well, that doesn't matter, because you've already stated in a previous post that slavery is productive because it produces something for the slaveholder.

Oh, that's not what you meant? You meant to say that slavery is productive because it produces the decreased likelihood of the slave being killed? Well, you playing League of Legends produces the same thing, but to a lesser extent, but it also produces the state of you not being bored!

Not being bored is not a product, you say? Well, Darren Tomlyn would disagree! He says: The nature of what is being produced does not matter at all.

Oh, but you meant that the REASON for undertaking the action is what matters? If you do it BECAUSE it is productive, it is work? Again, Darren Tomlyn would disagree with you, when he says: "The difference between productive and non-productive behaviour is whether or not the behaviour happens by itself, (because it's enjoyable), or whether it produces something."

Also, by your definition(s, because let's face it, you've offered many) any game that is supported by ad revenue can never be play, by your definition, as the action of you playing the game is productive.

Also, the playing of a game will likely result in you becoming better at the game, but also improving your cognitive abilities, which is productive. Now that you know that it is productive, no game can ever be play, unless you willfully become unaware of this fact (altering your perception). Again, this is YOUR definition: "if you know what you're doing is productive, then it's work, regardless of whether or not you enjoy it."

You say: "If you don't understand the word productive - talk to an economist. What productive relates to is work, and so it should be no surprise that matters dealing with work - the economy etc. - are where the main understanding and description of such terms can be found."

This is an embarrassingly circular argument. You imply: "Productive actions are those that relate to work. Work is an action that relates to productivity."

You say: "The difference between productive and non-productive behaviour is whether or not the behaviour happens by itself, (because it's enjoyable), or whether it produces something. The nature of what is produced DOES NOT MATTER (emphasis added)."

You've created a definition where every conceivable action is either play or work, which you already realized created problems when you explained that slavery was productive - it didn't matter that it produced nothing tangible for the slave (apart from avoiding death). You stated, clearly and unambiguously, that slavery was productive, because it produced something for the slaveholder. This means, unambiguously, that by your definition, that any single moment of game play that produces something of value to the game developer is WORK, if you are aware of the value being created.

This follows from your definition, not mine.

No one here is confused. You are. Stop pretending like you've understood something deep about the nature of work and play - your definitions are logically incoherent, they are not predictive or actionable, and you have yet shown a single concrete example of how not using your definition impacts game design.

Again, don't bother trying to say I've misunderstood you. I've taken the time to read what you say very carefully - perhaps more carefully than you wrote it - and it is very clear to me that you contradict yourself at every turn, present non sequiturs, tautologies, and logical absurdities.

Maybe you should take the advise of former commenters and go to college - maybe take a course in philosophy and formal logic, so you can avoid saying ludicrous and contradictory things.

Nils Pihl
profile image
Darren Tomlyn's argument, deconstructed:


1. All productive behavior is work. (Source quote: for any and all productive behaviour, is work.)

2. An activity can not be both work and play at the same time. (Source quote: Obviously work and play represent a very firm dichotomy – something done can only ever be one or the other, never both of any degree.)

3. All non-productive behavior is play. (Follows from 1 and 2)

4. The reason why someone is doing something is not what determines if the behavior is work or play, but rather, the productivity (outcome) of it is. (Source quote: They mistake what the word represents - in this case, of course, an application of behaviour - with why such behaviour takes place.)

5. If nothing is produced, the behavior is not productive. (Follows from 1-4)

Conclusion: Any behavior that failed to produce something was play.

Launching the Challenger space mission was play.
The Titanic was play.
Voting for John McCain was play.
The Civil War was play for the Confederacy.

Darren Tomlyn
profile image
5 is not a correct assumption.

I suggest you go away and learn what productive (and therefore non-productive) truly means in general before going any further - I'm trying my best here, but it's hard to explain to someone who really doesn't 'get' it.

As I said, it's all about perception. The reason that's the case, is that, (in bigger picture of human behaviour) very few things people do are ever truly non-productive - especially since people can learn from almost any and everything they do. The difference that really matters, then, is whether people perceive such behaviour as and by what is being produced, or not, (purely for the behaviour itself, because they enjoy it).

If your doing something purely for itself, (because you enjoy it), and no other reason - then it's play. If your doing something for any other reason, (usually related to an intended or actual outcome), even if you enjoy it aswell, then it's work. There is no other way of distinguishing between work and play that is consistent with how they are used.

That people, companies, organisations and things can FAIL in their productive endeavours, is neither here nor there for the definition and application of either work or play - if it's productive, then it's work, even if it fails.

(Of course, it's possible to produce something that is a toy - used for play - but this doesn't mean the act of creating it is also play.)

As such, none of the options presented at the bottom of your post involve play, (usually - though it's possible to vote for someone without caring about the outcome), since they all happened in order to (try and) gain a particular outcome - that they failed to reach the outcome they desired, has no bearing on the basic definitions of work and play, and therefore such applications of play to describe such things that happened would be inconsistent.

Nils Pihl
profile image
5 is not an assumption, it follows logically from 1-4. If you don't want 5, change 1-4. It's not a matter of opinion.

You are also contradicting yourself again. Please note that I quoted you on all assumptions that don't follow logically from other assumptions.

You're not arguing with me, at this point. You're arguing with yourself.

"If your doing something purely for itself, (because you enjoy it), and no other reason - then it's play. If your doing something for any other reason, (usually related to an intended or actual outcome), even if you enjoy it aswell, then it's work."

You'd be a lot more convincing in lecturing me on the English language if you could distinguish your/you're.

Look, I get what you're TRYING to say, but you're not saying it. Nestled in all the contradictions and incoherencies there's a decent argument - and we're in agreement on those points (when you're not disagreeing with yourself).

If you can't SEE that you're contradicting yourself, even after I've explicitly shown you how, I invite you to try to convince me again in an email - but I'm not giving you this platform to self-promote and be condescending to people who disagree with you.

I understand that you have a lot to gain by arguing with professionals at Gamasutra, and I admire that you've written somewhere in the order of 60000 words on the subject, but helping you win an argument against yourself in this thread is not that interesting to me.

I invite your email. Or add me on Skype ("nipibo") and you can try to set me straight in person.

Darren Tomlyn
profile image
The reason why I said that 5 is an assumption - (for you!) - is because it still needs to exist within the bigger picture, which you have failed to recognise, based on your arguments below it. (I didn't make the proper distinction, and also sorry about the typo :p - not sure if had time to proof-read it properly - (should know better than to post in a hurry, which is also probably why I didn't do a very good job. ;) ))

The problem you have, as I said, is in understanding the basic meaning of productive, in itself, in general, in relation to human behaviour as a whole.

What any behaviour produces, whether intended or otherwise, does not have to be tangible - which is why learning, by itself, is productive by its very nature.

It is for this reason why I said that most (but not all, certainly) human behaviour is productive, even if it's not always recognised and perceived as such.


So, there are three parts to this problem:

1) Recognising what being productive (and non-productive) means - though, as I said, I'm probably not the best person to explain this in the first place - (as you can probably already tell).

2) Recognising when something is productive or not for YOU, based on your own subjective opinion.

3) Recognising when something is productive for someone else, (even/especially if it's not for you), and vice-versa.


Generally, the act of creation is productive by its very nature, since it's about producing something - (we call them 'works' of art for a reason).

When it comes to producing objects, we don't usually have any problems recognising the difference between work and play, even if they're not intended to be used in a productive manner - (toys). (I don't think anyone would argue that being paid to create Lego etc. isn't a job.)

However, when it comes to creating things that are used to enable other, more abstract, behaviour - (activities) - we start to struggle.

The confusion between the words art and game is a basic symptom of this.

The act of creating a game, by its nature, is productive, and therefore work, regardless of the overall functionality and use of what is created.

This creative process, is what the word art itself represents, but from a particular perspective, in that everything we create tells a story (of its creation). If the function of what we create is consistent with this (telling a creative story), then it can be defined as being a work of art (produced by such a process).

Merely perceiving such a work of art, or even using it in an applicable manner, does not have to be productive in itself. Whether or not things are tools or toys, depends on the perception of those perceiving and using them.

Activities, being about behaviour in themselves, can also be productive or non-productive (take place for work or play) on behalf of those taking part or even perceiving such an activity. This is why a football game can be productive for those taking part, but non-productive for those watching. Again, however, such activities will still be productive on behalf of their creator(s).

However, we have an additional consideration to make on behalf of those taking part, which is not being recognised and understood, currently.

Although an activity can be recognised in general as being non-productive, (produces nothing recognisable and understood outside of the activity taking place), the perception and behaviour of those within the activity itself can differ.

Again, though, games can be productive or non-productive, in general, without any problems - (played (taken part in) for either work or play).

However, it is therefore possible that the behaviour found WITHIN an activity can be seen by those taking part to be productive for and within the activity itself, at that time, yet producing nothing independently, outside of such an activity, and therefore recognised as being play outside and/or after the activity has taken place.

Again, this all has to do with the innate subjective perception people have of their (and others) behaviour at any given moment in time.

Recognising and understanding this, is IMPERATIVE, for anyone wishing to create or enable such activities, and especially games - for the behaviour the word game represents an application of, places a lot of limits in how such different types of behaviour can be truly applied, and still remain a game.

So, how is this possible?

The answer to that probably deserves a post all to itself...

Zachary Pease
profile image
"If your doing something purely for itself, (because you enjoy it), and no other reason - then it's play. If your doing something for any other reason, (usually related to an intended or actual outcome), even if you enjoy it aswell, then it's work. There is no other way of distinguishing between work and play that is consistent with how they are used."

This part leaves me questioning what you think about goals in a game. For example, a goal explicitly stated in the game is play while a goal a player creates is work? I have a hard time believing that because the player could be perceiving their goal as productive, or they could just think it's funny or cool.

Part of why the argument causes so many slippery slopes is because you are arguing for a subjective interpretation of play. This inherently leads to questions of epistemology because we then need to agree upon how we know things. I feel that many people in Gamasutra do not want comments to go this route because it is never-ending, usually perceived as unproductive, and difficult to even gauge the conversations' progress. You would probably be much more welcome and less tense in a discussion about the philosophy of games. Nils Pihl is definitely trying to stay objective so he can give something tangible to the community.

In parting, I think you touch on something that is indeed a huge problem in gamification where expectations and perceived production can snuff out the sense of play gamification intends to bring to work. Namely in response to: "Again, this all has to do with the innate subjective perception people have of their (and others) behaviour at any given moment in time.

Recognising and understanding this, is IMPERATIVE, for anyone wishing to create or enable such activities, and especially games - for the behaviour the word game represents an application of, places a lot of limits in how such different types of behaviour can be truly applied, and still remain a game."

Alexander Ronalds
profile image
Always interesting to better understand the underlying psychology behind our games. Keep it up!

Ben Lewis-Evans
profile image
Hey Nils,

I am not sure I am that jazzed with the term "awards" instead of "rewards" (awards sounds a bit merit badgey and, for me, doesn't capture much intent) but at the same time I am not sure that it is an argument that is pragmatically important. I personally steal from economics a bit and use the term incentives (which captures the intent, to shape behaviour) and break them down into rewards and penalties. From a psychology perspective reward is also the dominant term in the literature, so it has that going for it (as well as being the dominant term in every day speech), but for me personally the fun argument is the distinction between a reward and a reinforcement and a penalty and a punishment.

Anyway, I will also throw my 2 cents into the play discussion. I think play is like baldness, hard to define exactly what it is, but you generally know it when you see it, particularly if it is happening to you. Intent and subjective experience is key, but I think being as precise as having to say play must be non-productive is overly limiting.

P.S. No real Dutch pronunciation this time, disappointing ;)

Nils Pihl
profile image
I understand that I am going against the grain when I choose not to call the incentive a reward, but I think it helps us avoid the paradox of rewards that are not rewarding. If you want to call "awards" something else, like incentives, that works for me - the main thing I want to avoid is terminology like: We made the game more rewarding by emphasizing points in the interface.

I am not trying, generally, to find definitions that match our every day speech - but rather try to stipulate what I'm talking about - there is no real reason to follow my terminology if you don't feel that it adds clarity. As I said, the main reason for stipulating terms is just to make sure we understand each other in this conversation.

The baldness thing is a hilarious analogy - and I submit to you that Gabe Zichermann is to game design, what Donald Trump is to hair ;) He can call it game design all he wants, but I'm not buying it.

Ben Lewis-Evans
profile image
The reward and rewarding distinction is not important one. Personally I think it can be maintained without changing terms and instead by stating "not all rewards are rewarding".

People are different and what rewards (or even awards) one may not another. So it is probably better to define these terms by the designers intent rather than the players reaction. If it is intended to reward, call it a reward. However, don't assume it will be rewarding to players (you have to check for for that)

Nils Pihl
profile image
I see what you're saying, but I disagree on what the best practice would be. I think stating that "not all rewards are rewarding" is insufficient to avoid confusion.

What would the definition of a reward be?

The way used here, the word "award" is what reflects the intent of the designer, whereas "reward" reflects how the award (that particular token) is received.

Either way, what I hope people take away from this video is that a something is only rewarding if it is a token of a currency that the player values. The logic of that is much more important than the words I happen to use, myself.

Ben Lewis-Evans
profile image
But that is the thing. Many definitions of reward are framed in the intent of the giver not the receiver. I am rewarding your reply by replying to you. E.g. I am recognising your feedback and by replying I am hoping to recognise your achievement of replying to me.

That you don't find it rewarding doesn't change that I was trying to give you a reward. So my reply is a reward if that is my intent, but it may not be rewarding.

I don't disagree with your core statement though. Which is just calling something a reward doesn't make it rewarding, and that semantic arguments are diverting but not much more :)

Arthur Hulsman
profile image
i read through all the comments, very interesting that so many words can be said when someone uses a word in a wrong context :D

Nils Pihl
profile image
My condolences. I owe you a beer.

Eric Salmon
profile image
Great read!

The "third person" in your example might be better understood as something like a finite (but incremental) currency game. I'm not good with the lingo, but what I mean is that the game tends to the infinite, but in finite steps (say, by introducing a 4th player identical to the 3rd):

3 passes 1 and 2, and stops.
4 passes 3, and stops.
3 starts again, passes 4, and stops again.

So really, the third is a combination of the first two--which might sometimes be more useful in analysis than just simplifying directly to an infinite currency game.

I mention it because if you define it in that way, your example currency games also reflect the top current business models for games: fixed price, free-to-play, and episodic--and give some interesting insights into their reward designs--and perhaps why players who are traditionally used to buying fixed-price games have been so hostile to the infinite pricing of free-to-play games--especially those who are motivated chiefly to "complete" the game.

On a different note, I think any game that wants to gain mass appeal really must address both the infinite and finite currency games. I haven't played the newer installments, but Call of Duty did this with both the single player campaign (finite), online multiplayer (infinite), and then also created a second reward system within the online multiplayer with unlockable equipment (finite), prestige (finite), and score(infinite)--and those are just broad systems outside of the actual gameplay.

Likewise, GTA and other sandbox games promote infinite play, but have discrete storyline/missions that can be completed.

Despite its popularity, Minecraft gets a lot of flak for being a a toy rather than a game--a criticism that after reading your article is mostly leveled at its reward system. It has several infinite reward systems, but until fairly recently didn't have a finite reward system at all--and even now it's a bit obscure.

Do you have any suggestions for identifying which type of currency game is preferred within a demographic? Or is designing to serve both usually the best approach?

Nils Pihl
profile image
Hey, thanks for the comment. In the video, I do refer to the third person as playing a finite currency game, in that it has a clear victory condition that ends the game. Having two people (3 & 4) play versus each other will continue until one no longer has any energy to continue.

That a game has an infinite currency doesn't mean it is infinitely fun! It just means that as long as you manage to keep the player engaged, there is no predictable end to the game. The next video will be on Engagement, and help elucidate that point.

Do I have any suggestions for identifying what types of currency is preferred? Yes.

It is not very easy within the current analytics paradigm (which is why my company is developing Traintracks), but what you want to do is try to earmark all actions within the game with potential currencies. Action A might be rewarding because of currency X, Y or Z, etc. Doing that will in many cases show you that certain currencies appear more often in the logs of engaged players. Look at your most engaged players, and look what their most commons actions have in common.

Quick example: In an F2P game that has IAPs including both functional and vanity items, you might find that your typical whale spends most of his money on vanity items. Next step is finding out what currency might be driving that behavior. Does he want to look "in fashion", does he want to look unique, does he want to express himself, etc. Studying other actions within the game can help make a qualified guess which one it is more likely to be. For example, if this player is also active on the forums or gloats a lot after winning a game, we might suspect that expressing himself is a key motivator. Increasing engagement for that demographic, then, could be done increasing the players' opportunities to express themselves.

Andrzej Marczewski
profile image
Great article. Your explanations are really well thought out. For me it is all about context and value (as I believe it is for you) when talking about what constitutes a reward. I think it is more a term of convenience that we use it in gamification. Most will not want to get into a debate about tokens and currency with a non technical CEO of a company :)

More importantly, your dissection of Darren was a thing of wonder, not laughed so much in ages ;-)