Gamasutra: The Art & Business of Making Gamesspacer
arrowPress Releases
September 22, 2014
PR Newswire
View All





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


 
Hierarchy of Needs for Rewards in Games
by Chelsey Webster on 09/23/13 08:59:00 am

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.

 

Introduction

Rewards are an important feature of any game, and as such, can be found everywhere across all genres. Rewards can come in any shape or size, and when given to the player appropriately can greatly increase their enjoyment of a game, motivating them to continue to play.

Rewards can even have a player do something that they don’t want to do, or don’t enjoy doing (which is quite opposite of the purpose of playing a game). Many players will do annoying, tedious or boring things for rewards. The ability to have a player do what they dislike of their own volition attests to their power. They are an essential tool to a designer.

Designing Reward Structures

Rewards shouldn’t be included on a whim; they must be scheduled and structured. The following diagram is a hierarchy of needs, adapted for reward structures.

Fig 1: Hierarchy of Rewards in Games

Based on Maslow’s Hierarchy of Needs, each section is a prerequisite to the one above, and can greatly enhance the one below. If a section is missing, however, those above it become ineffective. When designing a reward structure, it is important to work from the bottom of the hierarchy upwards to achieve a solid foundation. It is therefore important that each section is linked to the next. The higher up in the hierarchy, the more frequently the reward should be distributed; a fractal-like pattern should emerge (i.e. 10 minor rewards = 1 major reward, 10 major rewards = 1 core reward). It doesn't have to be strictly fractal of course, but minor rewards should considerably outnumber major rewards, and so on.

As we move upwards through the hierarchy, rewards become less intrinsic/more tangible. The bottom of the hierarchy is the most fundamental and intrinsic of rewards to the player – the player experience. At the other end of the scale, the top is purely cosmetic and cannot exist without the rewards listed below – however it works to enhance those below. Think of a scope for a rifle. The scope is great, and makes the gun much better; but it is completely worthless without the gun.

The following gives an overview of each section of the hierarchy and what kind of rewards may be found therein. Examples here will often be of the kind found in adventure or RPG games, but the principles can be applied to any genre. The rewards structure should always be designed in order of importance - from bottom to top.

Rewarding Player Experience:

Before considering anything else in detail, the player experience must be perfected. Good gameplay and immersion must first be obtained through the mechanics, aesthetics and UI (including essential player feedback). It is difficult to define whether this section is fulfilled, as fun is something that can only be experienced. The player must enjoy simply playing around with the game. The Grand Theft Auto franchise achieves this very well - who hasn’t switched it on, then ignored the missions and started wreaking havoc until they’re killed? There is no set goal to this, no rewards given (in fact the player is punished by losing ammo and paying hospital bills), but people do it because the gameplay/mechanics themselves are so much fun and intrinsically rewarding. Few games are fun to the extent that a person will play them with no prospect of rewards at all. Perhaps this is why GTA is so successful.

Core and Long Term Rewards:

Once the player experience is perfected, consider the core/long term rewards. These are very big rewards that the player will have forever. Examples include major plot developments, major milestones in opening new content (reaching a new island in Grand Theft Auto for example), or receiving a new game mechanic, such as those given after every boss fight in Crash Bandicoot 3. As these are the biggest tangible rewards that the player will receive and can affect the intrinsic experience, they should be given only occasionally. This retains their value and prevents annoying the player. People would become frustrated if they earned a new ability every 10 minutes; they'd never learn how to play.

Major and Mid Term Rewards:

These may be levelling up, completing quests, gaining large quantities of XP or money, etc. These rewards are significant to the player at the time but are not end rewards that are kept forever. They are very rewarding in themselves, but are also means to and end and will eventually be overwritten, replaced or become redundant.

Short Term and Minor Rewards:

These are small, frequent rewards which on their own do not benefit the player, but accumulate towards a bigger reward. Examples include collecting an item, small amounts of money/XP for defeating an enemy, or completing a stage in a mission. Just as Major rewards are means to a bigger end, Minor rewards are also means to an end on a smaller scale (the end being major/mid term rewards).

Cosmetic Rewards:

This is the final section to be considered. At the top of the hierarchy, they are purely cosmetic, serving only as a visual measure of the player’s progress. In some cases they are simply informative, such as numbers appearing on screen when you add to your score, or the squad selection screen in Mass Effect – the player can guess that they’re about halfway through when they’ve filled half of their squad slots.

On the other hand, visual progress elements such as unveiling a map or filling a bar can actively encourage players to do things that they would not do otherwise, as can achievements depending on design. A player might play for 10 minutes longer where they would have otherwise stopped because their XP bar shows them how close they are to levelling up. They may not attempt to round up and defeat 10 enemies with 1 grenade if there was not an achievement challenging them to do so, and may not just run around the entire area if they weren’t revealing their map.

While a game can still be enjoyable if the cosmetic rewards are removed, they will enhance the player experience as players like to see how they're doing. Visual rewards are vital in helping the player to gauge and anticipate rewards which is an encouraging motivator. Visual rewards never actually affect progress or any other aspect of the game at all.

There is a blurred line between the player feedback (mentioned at the bottom of the hierarchy), and cosmetic rewards (at the top). Player feedback is essential communication to the player, but cosmetic rewards can exist on their own (as with achievements, etc), or tie into player feedback for visual enrichment. One big difference between the two is that omitting essential player feedback can be game breaking, or at least break immersion. An example of essential feedback is seeing your avatar animate as you move around in the world. if it simply floated across the landscape this would completely break immersion and make it unplayable to many players. A problem like this is found to an extent in Oblivion and Fallout 3 (3rd person). The avatar animates but not it doesn't match its direction if running diagonally, so the character appears to float around. When playing in 3rd person this can completely break immersion. You can see an example of this here.

While it should be designed bottom up, the player will often experience much of this hierarchy from a top down perspective. They will experience cosmetic elements and minor rewards to begin, such as obtaining a task (this is where they are initially given the prospect of rewards), then minor rewards such as collecting items and XP, which lead to major rewards earned by completing the quest or levelling up, and eventually core rewards such as a new mechanic or unlocking a new world.

Application to a successful game:

This hierarchy includes rewards featured in World of Warcraft, a very popular (and notoriously addictive) game.

Fig 2: A Basic reward structure of World of Warcraft

(does not include all rewards)

These are just a selection of the rewards offered, but it is clear that the game is abundant in rewards on every level. The world and the gameplay are immersive and such high customisation of the player experience (classes, specialisations, PVP/PvE, RP, and so on) means there are hundreds of player experience combinations in one game.

Add to this the limitless supply of rewards of all kinds which all tie in to one another and accumulate towards something. Then, to top it off it is peppered with random chance rewards. Even the most mundane creature can drop disproportionately powerful rewards on very rare occasions. There is always a reason to keep playing; this is clearly a very strong reward structure in every aspect.

Application to an unsuccessful game:

As part of an experiment for my dissertation I created a very simple 2D space shooter game. Though it was fit for the purpose of an experiment on rewards and yielded conclusive results, the game would not be popular or successful in the real world. After developing this hierarchy I decided to apply it to my game:

 

Fig 3: Reward structure of a basic 2D space shooter

The game is strong at the top and gets weaker as we get to the fundamentals of the hierarchy. There is no hope for this game on the market today no matter how good the upper sections are, because the lower sections are not fulfilled. As well as gameplay being lacklustre, an entire section is missing. Tight time and resource limitations meant that the game focused on tangible rewards which could be created quickly and easily. There was no time to invest in excellent gameplay; it just had to work.

It is easy to draw a comparison between this game and popular arcade games from the 70's. These were successful in their day, though they did include the long term reward of top slot on the leaderboards - but today their main appeal is arguably nostalgia. Todays games have far more sophisticated gameplay and reward structures, so a game like this cannot compete.

Consider the World of Warcraft hierarchy if a section is missing:

Fig 4: World of Warcraft’s reward structure with a section removed.

With this section missing, the hierarchy doesn’t make much sense. Level increments are gone, for example, so now the player must go from level 1 to level 90 (or whatever its new name would be) in one grind. The player would be consistently weak and the suddenly immensely powerful, but only after an inordinate amount of time. Needing  999,000,000,000 XP to reach the next/last level is far too discouraging when the player is getting 20XP per enemy, even more so when their XP bar never seems to budge. Bitesized chunks are essential.

The removal of this section has completely devalued those above. The minor and visual rewards contribute so little to the next goal that they don’t feel very rewarding at all. Players are much happier and more likely to play if they feel that their goal is realistically obtainable.

Summary

Quality does not equal quantity, but it is important to have at the very least one reward in each section. Aim to fulfil each section of this hierarchy to a high standard and ensure that they accumulate towards one another. Though it has not been around long enough to be tried and tested on a large scale, following this structure should achieve a strong, structured, rewarding player experience.

While many of the examples throughout this article are based on action adventure/RPG type games, they can be applied across any genre, from puzzle games to sports and racing. By applying this model when designing a reward structure, the designer can ensure that the game has solid foundations and is rewarding enough to make the player feel good. This can ultimately keep them playing the game. 

Though I've tested theory this on multiple games and have developed it over time, I am always keen to improve. If you have any suggestions feel free to let me know (:

Thanks for reading!

 

 

 

 

 


Related Jobs

Retro Studios - Nintendo
Retro Studios - Nintendo — Austin, Texas, United States
[09.22.14]

RETRO STUDIOS - Level 3 Engineer
Blizzard Entertainment
Blizzard Entertainment — Irvine, California, United States
[09.19.14]

Art Director - World of Warcraft
SAE Institute
SAE Institute — San Jose, California, United States
[09.19.14]

User Interface Design Instructor
SAE Institute
SAE Institute — San Jose, California, United States
[09.19.14]

Compositing Instructor






Comments


RJ McManus
profile image
Hey Chelsey, like you I'm a recent graduate looking to get into game design (and as such my opinions are by no means those of an expert), but my academic background is in history and- slightly more relevantly- psychology, so I thought I'd share some feedback.

The first thing I noticed was that in one sense your theory is a subversion- or perhaps an inversion to be more precise- of Maslow's Hierarchy of Needs, which is pretty cool. By that, I mean that you place the "big picture" of player experience in a supporting role at the bottom of the pyramid, whereas the "big picture" of self-fulfillment rests at the top of the pyramid in Maslow's model. I assume that you do this intentionally to make a prescriptive statement about the superficial importance of the "Cosmetic Rewards", and I would be inclined to agree that more superficial rewards cannot form the core of a profoundly satisfying player experience. That said, it would be quite interesting to hear your thoughts about this perspective, and whether this was an intentional move on your part.

However, I think there are some examples of games that have been successful without ever establishing the lower tiers of your reward structure, even if you and I would criticize their design. Indeed, sometimes it seems as if the promise of short-term feedback is enough to fuel players forward, regardless of whether the experience is intrinsically satisfying (i.e. "grinding"). I believe this distinction between intrinsic and extrinsic rewards is a crucial one, even if it is ultimately artificial in nature, and your theory seems to flirt with this idea. I have been working on a post addressing this distinction in the context of "immersion" and "flow"- which seem to be every designer's favorite (or un-favorite) buzzwords. I have not finished it yet, but the gist is that immersion serves to optimize the player's enjoyment of intrinsically motivated actions, whereas flow optimizes enjoyment of extrinsically motivated tasks.

At any rate, contemporary psychology has in fact largely dismissed Maslow's Hierarchy of Needs under the basis that our "needs" need not be fulfilled in any particular order, but one can certainly make a case for it nevertheless applying to games. I guess that I should assume that you are already familiar with the MDA framework since you are a former student of game design, but I think it makes an interesting observation that top-down designers and players may experience the levels of hierarchies from opposite directions. Theoretically, the designer envisions the big picture and extrapolates outward, whereas the player experiences the big picture only as the sum of the finer details. Your model seems geared toward a designer's perspective, but I would be interested in hearing what you have to say about this possibility.

The other main insight I have is that rewards can be thought of as simply one type of feedback (along with punishments), and perhaps then both positive and negative feedback are equally important to the player's experience. It would probably be a very interesting thought experiment for you to consider whether your theory accommodates both positive and negative reinforcement, or whether it requires any modifications in order to do so. Oh yeah, one more thing (though I forgot where I read this): I've seen it argued that even feedback on the level of watching your avatar perform an action as a result of your pressing a button can be thought of as a reward. Do you agree with that or have any thoughts about where it would fit into your model?

Last but not least, I'd like to say that you're an excellent writer and I look forward to reading more about your ideas in the future! And sorry for the TL;DR!

Chelsey Webster
profile image
Hi Ryan, thanks very much for the comments and feedback, it was an interesting read (:

The hierarchy was indeed designed with the big picture at the bottom, though I found it to be both similar and at the same time different to the Survival section of Maslow’s hierarchy. I’d considered it similar in the sense that gameplay/UI/UX are the lifeblood of any game, and that nothing else would be important without it (we can’t enjoy our nice cars and popularity if we’re dead, after all!). On the other hand, it is very different in that survival is not at all a fulfilling or enjoyable way of life, but a good player experience is quite the opposite. So while I didn't begin designing it with the goal of subversion, I did so because it made sense and felt more intuitive. It was certainly an initial aim to state that superficial rewards cannot form the core of an experience; that they can enhance an already good experience but cannot carry an otherwise poor game alone.

Could you give any examples of games that are successful without the lower tiers? I’m sure there’ll be the odd exception out there but I haven’t come across one yet and I would like to find some. I have created this with the designer in mind too, yeah – I thought that it could be of use to some designers if they’re unsure on how to reward the player appropriately, and following it may help designers to ensure that nothing is missed out, and that their priorities are correct.

I had thought about things such as avatar actions as rewards, and was considering including it but I thought that it might sound too pedantic :p I wasn’t aware that other people had made that argument but I’d thought about it – so it already has a place in the hierarchy yes (: This would be placed in the bottom of the hierarchy which includes basic player feedback. I’d put it here as its feedback for essential information rather than a cosmetic reward which doesn’t *need* to be there but is nice to have. The biggest difference between these I think is that the player will definitely notice the lack of essential feedback as it would be very jarring, say if their character just floated across the world rather than running (oblivion/fallout 3 avatars did something like this and it really broke immersion for me). Cosmetic rewards can only add to a game, but missing essential feedback will detract from it. (I'll add this to my blog post in fact)

I’d looked into punishment for my dissertation briefly when looking at the Skinner Box experiment. I hadn’t considered incorporating it into this, but it is an interesting idea – I will definitely have a think about that!

RJ McManus
profile image
Regarding the ordering of the hierarchy, I can definitely agree with that line of thinking, though it'd be interesting to see how someone would go about arguing against it (you see a lot of unconventional and interesting opinions here on Gamasutra).

To me, the mere prevalence of the term "grinding" (most commonly in the context of MMO's but not exclusively so) would suggest that there are very many games in which the core player experience is not rewarding at some fundamental level, and yet the player still finds him/her-self motivated to continue playing nevertheless. Which games constitute a "grind" probably varies significantly between individuals, but for me Temple Run and WoW would both elicit a sense that I am playing simply to achieve some far-off reward and not because the process is particularly fun.

I suppose that your model could respond to this criticism in a number of ways. Perhaps players referring to games as "grinds" simply indicates that the genre doesn't appeal to them, which has nothing to do with how well the reward structure has been designed. Or maybe the fact that players are inclined to keep "grinding" indicates that the core experience is in fact rewarding, but simply in a subconscious way. That said, perhaps players are capable of bridging one empty level of rewards if they are sufficiently externally motivated to do so.

Another idea I had is that if you really look at the essence of the rewards in the middle tiers of the hierarchy, their significance ultimately tends to fall under two kinds (rather than being of any distinct type). While leveling up is definitely a longer-term reward than gaining some small amount of experience, in some sense it can be just as superficial in nature (in either case some number has been changed), but leveling can also (sometimes but not always) contribute to the core player experience by unlocking new options for the player. Thus you have a gradient between the "essential rewards" at the bottom of the pyramid and the "cosmetic rewards" at the top; this is another nuance that you may have intentionally embedded in the theory.

Indeed, my first reaction to the argument that avatar feedback constitutes rewards for the player was "how pedantic", but it's an interesting idea, and your theory's accommodation of it makes sense to me.

The distinction between rewards and punishments (in their general sense) as subcategories of feedback is ultimately quite subjective; the computer systems that run games arguably cannot make such value judgments and thus they see only output/"feedback". Thus, I do think that it would add a lot to the explanatory power of your theory if you figured out a nifty way to accommodate both rewards and punishments in your model.

Kenneth Poirier
profile image
This is awesome. Thank you.

Roberto Dillon
profile image
Wouldn't it make more sense to order the rewards in terms of frequency? i.e. "Rewarding Player Experience" stays at the bottom, then the "cosmetic rewards" (visually informing the player about her progress should actually make her feel safer, which matches exactly Maslow's second stage in his pyramid), then "minor and short term" followed by"major and mid term" and the "core and long term" at the top of the hierarchy ("self actualization" for Maslow, i.e. we empower the player and give her more abilities and skills to play with)?
Just my 2 cents.


none
 
Comment: