The Design of Time: Understanding Human Attention and Economies of Engagement

[Music] so just start so my name is Chelsea how I spent the last eight years in the free-to-play space doing a bunch of different projects and I'm now working at Alchemy labs in VR and one of the things that was consistent throughout my time in freeto playay was just that this idea of time uh it's the underlying foundation for a lot of free-to-play mechanics and yet one thing that I noticed was that people don't usually think about time over time we think a lot about short-term gains little spikes but we don't really understand human attention

and behavior past kind of the daily to weekly stage and I think one of the major things that I saw was this tendency to underestimate how quickly humans adapt to stimulus which is just a fancy way of saying people get bored and as we see time becoming more foundational and things Beyond freeto playay in console and AAA I think it's really vital to look at human behavior in Psychology and understand how exper experiences can really engage people long term after all the Holy Grail would kind of be a game that lasts and is lucrative forever

dep independent of developer content I don't know how to do that if I did I wouldn't be here but here's some pillars so I want to start with player engagement over time now when I talk about player engagement I'm talking about any I'm not talking about any discrete metrics so no da or ma or some of the other things that we've used uh in the past when I talk about player engagement I'm talking a little bit more abstractly about the time attention and mental effort a player gives to a game regardless of whether or not

they're playing it so essentially the psychological energy they're putting towards your game at any given point in time and I want to start by thinking about this through the lens of another area of Our Lives where we use the term engagement after all marriage is supposed to be this thing that lasts for life even if statistically we haven't been doing a super great job of that with the lens of relationship psychology is still really helpful as we think about our games after all historically Developers wanted to see players fall in love with them to really

grab them and get them engrossed and for each day to be better and more exciting than the one before um and we know that that's not really how relationships work but love in the short term does start with infatuation and this is the typical Peak we see when players first pick up a game we see them binge we see longer player sessions more frequent sessions per day but this phase of Engagement is inherently transient the getting to know you roller coaster never lasts because we as humans are designed to adapt and find patterns we're designed

to get bored this is how our brain optimizes to live in our world even the most beautiful rock in the world is going to become background noise if it's on your finger every day and that's just a result of how we as humans process stimulus at the most basic level whenever we receive a stimulus we tend to pass it through the list of our basic needs can I eat it mate with it or will it kill me once we've done that our engagement starts to fall off we run it against some of our higher order

cogn itive needs and I want to point out that uh this graph is not accurate the peak and amplitude the rate of decay all of that is dependent on different personal factors and factors with the stimulus and content itself I'm going to use those two words stimulus and content relatively interchangeably any piece of content is going to be a stimulus that evokes some form of player attention and engagement now the other half of this curve if we know that the stimulus is coming up is this idea of anticipation this essentially means that we get engagement

before we actually even give the players something now I mentioned that it doesn't usually look like a bell curve at all but the one thing regardless of that decay rate and amplitude or the rise of the anticipation is that it's going to go down over time it will Decay and this is just a natural result of some of our neurological Tendencies so here's a few different perception based reasons for this this is a small small subset if you're interested in this I highly recommend looking up a few basic ux books on principles of human attention

so the first thing that I want to talk about is that we're hardwired to find patterns in the world as soon as we find a pattern it means our brain doesn't have to do so much work and we can attend to other things once we find a pattern we only pay attention to the way that some sort of stimulus deviates from it this also means that as soon as we recognize a piece of content that follows a pattern we're not going to get the full amount of Engagement we normally would we saw this a lot

early on in so-called golden mechanics it turns out that golden mechanics were really just easily repeatable patterns and the initial spikes of Engagement that we got from them didn't hold up over time the next thing we do is that we chunk things the iconic example of this in ux is the idea of phone numbers it's much easier to remember three chunks of numbers than 10 individual numbers however this also means that we don't get 10 individual spikes of stimulus nor the sum of 10 individual parts we get something a little bit less than that um

you can see this in games in something like Pokemon where when I spin a specific Poké stop I don't get an individual Spike of Engagement for each item that drops I just get this Collective chunked idea of a pokey Spin and I think the biggest one is that we tend to normalize put simply everyone has a different base emotional level and that base emotional level isn't what determines how we feel what determines how we feel on any given day is the way that the stimulus we're exposed to deviates from that Baseline so on the plus

side this means that no matter how great or terrible your life is on any given day designed to exist and survive to continue moving forward so just being able to feed your family might feel as good as someone with a relatively no experience a low experiential Baseline as closing a billion dollar acquisition for someone with a relatively High experiential Baseline now unfortunately if we continue to give people these spikes that doesn't mean that we're going to give them bigger and bigger emotions what it means is that we're going to elevate their Baseline so essentially just

like in Adrenaline Junkies the first high that we get is is never going to be the same okay so that sounds a little bit dire but I want to look again at relationships we know that some of them have to work after all we still exist so what does that transition look like how do we get players from that initial infatuation to a place where they're willing to be more intimate and committed the first thing we need to do is not panic when that initial Spike wears off in the early days of getting Mass data

we saw this all the time we'd see the slightest little dip in engagement and we'd set off all of the alarms would get called in the middle of the night the end result was that we were constantly trying to shove bigger better grander things at players all the time fearing any micro dip in their attention now the result for players was exhausting we wound up making our average players feel like they were missing out and driving our highly engag players into this unsustainable frenzy and the downside is that once you start on this path there's

really not a lot of options beside continuing you've elevated the expectations for your highly engaged players youve basically trained them to constantly want more at the same time your average players start seeing a higher and higher gap between their experience and the experience of these highly engaged players they start seeing fewer and fewer reasons to stick around so instead of more more more more more what we have to do is start looking at ways to say enough to our players instead of striving for this ever increasing engagement we need to get them into a habituated

series of cycles of Engagement in the same way that we need rest in order to be able to act we need relief between these periods of attentional arousal so there's a lot of really common ways to do this in a game like choices you can read whatever you want but you can only read a chapter every few hours and something like wow you can battle your pets all the time but on weekends it's going to be faster and more rewarding and in something like Clash Royale you can continue to attack as often as you want

but that high reward the chests are inherently limited and I want to talk about that one specifically because I remember being in design meetings and hearing other designers say myself included actually I can't believe how blatant this chess limit is they're not even trying to hide it anymore and I want to talk about why that's brilliant because another Cornerstone of healthy relationships that are designed to last is trust and consent now I know that those are heavy words but we as developers really want engagement after all we live in an engagement economy attention is essentially

our coin so we really really want those players to like us so wey to get their attention and we don't always do it in the healthiest of ways now a lot of us are familiar with M mahai apologies uh for his work on Flow but he also did a lot of work on taxonomies of attention he said that there are essentially two categories of attention one is voluntary we choose to attend and focus on something and the other one is compulsory which means our attention is essentially demanded by some sort of perceived emotional biological or

social need now I want to point out that by my definition compelling attention is non-consensual and for that reason compelling attention will always perform better in your ab tests of course forcing someone to do something is going to get you the result you want more than letting them choose to do it and this is why we see these techniques all the time in games it's why we see them all the time on TV and it's why we see them all the time in our social feeds and this is repulsive this is using sensationalism and implied

risk to force us to attend to something this should infuriate all of us this diminishes our agency and our autonomy as humans it those are the very things that make us human but the data the promise of profit is not permission for unethical action if you're going to treat data as permission to act immorally and unethically to put your business ahead of the people that you are supposed to be serving then I want you all to think about what the future is going to look like in this datadriven world if you use data as an

excuse as a way to alleviate your own responsibility in a situation that isn't creating a relationship this is 21st century enslavement we no longer need need to hook people up to get their physical labor but we live in an engagement economy and we are stealing people's attention so I don't care what your data says data is a manifestation of action it's a means of tracking behavior and data is oblivious to The Human Condition just because you can do something just because something would make you more money doesn't mean that you should do it because regardless

of how ubiquitous these techniques are they are fundamentally dark design all dark design does is slowly erode trust and faith in people's autonomy over their time and when we reduce people to just those animalistic impulses those things that we cannot control we are reducing people to animals we are removing their personhood and we're already starting to see the horrific results of this as our own capacity to focus our attention is lost to this constant barrage of alarm and fear in media we are sinking into a world ruled by those fears if you were at any

of the ux talks these past two days one consistent theme was that when we are afraid we do not grow we do not learn and we do not flourish so think about what that future looks like but this was supposed to be about games so let's go back to clash Royale everyone was right they aren't trying to hide anything and in doing so they're actually giving players more control over whether or not they choose to attend to a game they're letting players decide whether or not engagement is worth it instead of confiscating and disguising their

demands for time when we're clear with our boundaries and limitations we're actually being more respectful we're being less deceitful more upfront and allowing informed negotiation that is what builds trust and faith over time even if it can be harder to stomach at first so let's talk about that harder to stomach let's talk about player perception of time now the role of time in freeo playay games is essentially a pacing tool it controls the consumption of content like any temporal art pacing is really vital to a well-designed experience but for games that depend heavily on developer

created content it's also one of the only ways to survive which means we have a very large capacity to try to get uh that pacing in ways that aren't always healthy so how do we surface time in a way that perers perceive positively how do we integrate it into that healthy relationship we want to create there's a few different techniques for surfacing time there's direct timers and that's exactly what it sounds like a little 2hour countdown there's indirect uh inclusion of time which is where something might take 200 mana and you generate 100 Mana per

hour again those are functionally identical they both delay your consumption of content for 2 hours but they feel and can feel very different one thing that affects the way a timer feels is ver similitude and context ver similitude is a fancy way of saying does it fit with an existing player model if you have something like time uh crops needing time to grow that fits with our existing mental model it's going to feel better than something like Keys randomly sprouting every 3 hours the other difference between direct and indirect timers is that indirect timers actually

let us have some autonomy over our time Equity we build up that 200 mana and we don't necessarily have to use it on one specific thing we can use that investment of time that we have whatever way we want the final way to surface time is implicitly so this means that some mechanic in your game literally takes time to complete so something like just walking around and catching Pokémon and Pokémon go takes time it takes time to move your body from one location to another and this used to be considered the golden uh kind of

the golden standard and console era but what it means is that we don't get any benefit from a game unless we are playing full paying full and complete attention to it and that um doesn't always feel good either so essentially what I'm saying is what I'm saying is that none of these are specifically right it's all about how you use them not which one you use and some of the most successful games ones that feel the best utilize all of these in different ways but because direct timers tend to get that unusually bad uh reputation

like they did in Clash I want to point out another benefit of being completely upfront with how much time you have until your next piece of stimulus that's the creation of Rhythm and ritual essentially if we're told every day that we can come back and get a little um a little bit of new content that's going to help create that habituated uh cycle of Engagement that's what we want and then ideally what we do is we layer those over time so we create pieces of content at different time intervals and as those stack you wind

up with a steadily increasing curve that looks a lot like a flow curve this is awesome so the things that I mentioned previously were all usually either momentto moment session or daily types of stimulus so I wanted to just point out a few examples of people who are doing longer time scale rituals relatively well in Summoner War you have and a bunch of other games you have things that change every day so if you need a specific resource you can say ah I need to come back on Wednesday in game like Hollywood University leveling up

characters is required to uh unlock new content but leveling up a character can take up to two weeks at later levels in games like ingis or Neopets which has been around for well over a decade you have different badges in forms of social signifiers that can take from half a year in this case to over a year in the case of Neopets look at that you just have to have a snow bunny for a year then you have games like Fallen London where rewards can stack over multiple years so when you come back the next

year you actually get something that wasn't available the first year and taking this to its most extreme you can look at something like Evon line where the total time to complete all skills is estimated between 18 and 25 years all right so 25 years is a really long time uh it's not something that we tend to talk about I've never heard someone look at that time scale in a design discussion unless I was the one instigating it um by imagining a long-term relationship with players we can start surfacing little things that help Prime them for

the idea that maybe this game is worth more than just a few days or a week or a month of their time if they go in assuming something is going to be a quick fling that's probably going to be what their behavior is so I said the word surfacing so I want to point out that no matter how many of these rhythms you have if you don't surface them people can't anticipate them and they've done studies have shown that people who take pictures of their food actually enjoy that food more specifically because they have that

moment of appreciation and anticipation I don't know if you any of you guys remember what Christmas Eve was like as a small child but I was pretty sure it was going to last until I died so as you design remember to give them the ability to anticipate remember to surface things so that players know they exist and remember to do it in a way that's not really creepy if you start talking about being buried next to each other on the first date that might be a bit of a turnoff okay so how do you make

anticipation a little bit more sustainable I talked about how Variety in a pattern grabs our attention variability does the same thing for our anticipation and that's essentially because we're optimists if I say imagine a plate of spaghetti you're probably not imagining a few dry cracked strands on a plate you're probably imagining that optimal version of spaghetti you probably imagine a heaping plate with a little bit of sauce and depending on what you like maybe a few meatballs so when there's variability and we know know the potential outcomes we tend to aim for the uh the

best one even if that is very very very rare now of course this has huge opportunities for abuse so again this is something that you can't just Implement blindly you'll see one of the themes about this whole talk is actually taking responsibility for the decisions you make as designers and thinking about their impact on players and whether or not that's in the player's best interest okay so there's a lot of different ways to add variability one of the most effective quote unquote is variable ratio which means you tell people they're going to get something after

a certain number of actions and you don't tell them how many actions that is and it always changes now that leads to the most engagement which means players will do that action again and again and again and again uh they'll binge right so that's the same behaviors we talked about earlier they'll binge and they'll burn out now that doesn't mean that variable ratio is bad it means that perhaps if you use variable ratio mechanics should be for a weekend Blitz event or one day a week you get this variable ratio because those spikes of Engagement

are enjoyable when you hit your 20th anniversary and Valentine's Day comes around that can be a fun time to spike your engagement to do things you don't normally do so again none of these things are inherently bad but you need to think very consciously about the effect on the overall player experience if you look at the worst implementation of variable reward you get fixed interval which essentially means you tell people when they're going to get a reward and and they tend not to engage as much until that time comes around that is something that is

much gentler and something that could work as a through line that's constantly present in your game okay so you do all of these wow we are running out of time all right I'm going to switch this into quick mode so you have this solid uh Cadence of contents and events it starts looking like a flow curve but even if we just give people a tiny little bit of stimulus they're still going to wind up with this huge horde of it and that gets into all sorts of problems once you add social because your Elder players

are you can't ever catch up to them as a new player and that leads to all sorts of problems in this healthy kind of social ecosystem that I'll talk about a little bit later so let's how talk about how we can take that progress and turn it from a something that's completely linear into something that has a little bit more Loops this is my beautiful Art thank you um the first one is Decay so Decay is a form of Maintenance which means a stat goes down over time and must be continuously boosted this is used

in Kim Kardashian where if you don't come back to the game your partner will leave you it is a horrific experience especially because they ask you for premium currency to resolve this but again the mechanic is not negative by default and something like Neopets they use ver similitude and context just like I mentioned before to make it feel better we're used to in the real world pets needing food so of course it makes sense that I need to consistently feed my pet durability is basically user-driven Decay this tends to feel a little bit better because

instead of Decay being driven by time it's driven by things that we do as players a great example is Tools in Minecraft the next one is bounded Decay so this essentially means that a stat goes down over time up to a point this was something that we saw a lot in Old RPGs progress was getting from point A to point B once you hit a save point the progress you'd made towards point B couldn't be revoked only the uh only the progress between that save point and the next one could the next type of Decay

is Market driven Decay um you see this a lot in Stark stock markets and style what this means is that the actual value the actual kind of properties of an item don't change but the value changes because of the perspective of others so you could have the hottest jacket in the world one week and then the next week it's out of style the color is wrong and the cut doesn't look so good anymore next type of Decay is other driven Decay so this is basically Zero Sum I think you all know about this and the

final type of Decay that I want to talk about is mandatory Decay so we're seeing this a lot in idle games where you actually have to give up all of the progress that you've made in order to continue um you do get a little bit of metac currency at the end which then fuels to your next run through the game the interesting thing about this is it gives players control over the pacing of their own experience in a really valuable way eventually as you're moving through that First Progress bar in an idle game the return

on your time starts to get worse and worse and worse you decide when it's no longer worth it and when you do you exchange all of that progress you've got given for that new meta currency that also Shucks you back to the beginning of the progress bar which as I mentioned before tends to be that period where we get much higher volume of stimulus it feels really good I also want to point out that Idol games have some of the best retention in the single player Market it's a good example of how when we actually

give autonomy back to players they reward us with their attention over time okay and progress is just one form of satisfaction uh there's lots of different taxonomies of satisfaction Perma ocean self-determination Theory but progress is always a part of those um theories and the other part that's always consistent is this idea of relatedness so I want to talk uh finally about communities over time any game that has lasted for more than a few years has relied on either algorithmic content like Solitaire or social content and whether that's a co-op game or an MMO or something

as Timeless as Sports the content the stimulus doesn't come from the basic systems it comes from the fact that other people are using them each individual in the world is the this infinite well of potential content and once you can harness that you've got the ability to generate content forever it's very exciting so how do we make relationships form there was a great uh report out of project horseshoe this year that broke down some of the fundamental pillars of friendship I highly recommend that you read it it's based on some psychology studies from the 1950s

that have held up over time which is always exciting the first thing that you need is proximity so this means people need to see each other free frequently so if you've got millions and millions of players you need to find a way to Shard or subsection your player set so that people encounter each other again and again when people start to recognize each other that builds Affinity the next thing is similarity so this means I need some form of self-expression that other people can relate to this can be as simple as something like team Mystic

or Valor and Pokemon and it goes all the way down to kind of complex uh Badges and signaling tools and uh user customization the next thing is reciprocity so we need to be able to form a symbiotic relationship which means I can do something for you you can do something for me this doesn't me to be item exchanges this is something as simple as a chat window I don't know how excited you guys get when you're talking to someone that you love or respect and admire and you see the little dots in Gchat start to

move that's very exciting that's a form of anticipation and stimulus and then the final pillar is this idea of disclosure which essentially means that I can take risks with you I can express more of myself with you and those are some of the underpinings for things like trust and compassion and empathy which should be familiar as things that create longer term relationships now before we can do any of those we have to be able to communicate essentially if people can't communicate their ideas and their feelings in your game they're not going to be able to

build complex relationships it doesn't matter if this is private messages message boards chat rooms the ability to communicate is a fundamental necessity and if you can get that into your game the less people are going to turn to outside Services which means the more your game becomes less of an activity and more of a place place to hang out relax and exist together once you've got that underlying Foundation you can start to look for signs of societies and it turns out they look a lot like real life you have social activities cultural activities economic uh

activities and the great news is that you don't need to build incredibly complex systems in order to evoke some of these social behaviors we as social creatures inherently tend to produce these even with basic systems as soon as you have a power curve for instance you can start getting different political um elements if I more powerful than you I can exert some sort of will or Force I can lead you as soon as you get public spaces and the ability for self-expression you start getting cultural elements Aesthetics uh Fashions if people can Proclaim things in

chat you start getting morality and belief systems as soon as you have item exchange you start getting economies you can do charity banking investment and as soon as you have some form of player customization you start getting things like group identities specializations celebrities that you recognize and these are all really really cool things but what's that kind of through line if we look at player customization and I hope that the I hope that it's bright enough here um what this Guild has done is they've all worn a black tabard with some yellow pixels on it

but those aren't just yellow pixels for this group those yellow pixels imply a shared history a shared set of values shared trust and understanding those yellow pixels have meaning meaning means that an item has greater value or greater importance than what the raw materials imply meaning is psychological value not physical value in the same way that attention is our psychological energy as opposed to our physical energy so to create meaning players must invest time and attention into something and if they do that together it works as a sort of multiplier that's why we have religion

and why we love our countries sometimes and so if we want attention one of the best and healthiest ways that we can get it is to enable social structures that generate meaning this creates this beautiful upward spiral meaning begets attention which generates more meaning and once you have that escalation you have the best underlying foundation for increased engagement over time meaning enriches life it's this innate human need and the more we can facilitate it the longer our product is going to last this is the reason that Facebook is still going we fill it with meaning

we fill it with our friends and our photographs our idle chatter we fill it with memories and little anecdotes we fill it with meaning and that investment is very hard to leave behind when we give something our time and attention when we create meaning together with someone and when we're treated with respect and love and return that's the true recipe for long-term relationships so please be good earn the attention of your players don't demand it be trustworthy y communicate clearly and transparently with your players find a balance in the relationship that works for both of

you be farsighted don't follow these short-term spikes and don't don't believe in the long-term success of malicious interaction patterns plan a journey that's going to last for years and for the love of God please do better as we design games that can last for years we need to migrate from being storytellers and content creators into Engineers of new societies societies where people trust each other or at least know the risks of wandering into wild territory uh societies where people can play together and challenge each other and unite together to form these complex social structures we

need to be Architects city planners psychologists and anthropologists and we need to look towards the horizon instead of staring at our feet we live in an engagement economy whether we like that or not attention is our coin so please be mindful of where you spend yours and what you're incentivizing when you spend your attention you are literally paying attention and people are going to use that value that they get from you and please be honorable to those that you ask for their attention our society right now is a reflection of the systems that we've created

so go and create systems that are better create systems that are going to help our world Thrive not just right now but over time thank you [Applause] [Music]