Using Variable Rewards
Now that we've talked about the action phase and we've talked about how to make that behavior as simple as possible to make that that key action is easy as we possibly can let's move on to the next step of the hook the reward phase when we talk about rewards, we need to start in the brain and in particular an area of the brain called the nucleus accumbens, which was first studied by two canadian researchers by name of olden milner, who in the nineteen forties implanted the brains of lab animals with these tiny electrodes and they gave these lab animals a little lever to push. And every time they pushed on this lever, the lab animal would receive a tiny electrical jolt to this part of the brain to their nucleus accumbens and what old and milner observed wass that the's lab animals stimulated this part of the brain incessantly they would forgo food and water they would run across painful electrified grids just to continue to stimulate their nucleus incumbents in later experiments done on...
people the researchers observed similar results that when people's brains were implanted with these electrodes and they were given a tiny button to push on to send electrical jolt to this part of the brain they did so hundreds of times some people in the studies had to have the machines forcibly removed from them to get them to stop activating these buttons now it turns out we don't actually need electrodes to activate our nucleus accumbens because your nucleus accumbens is activated every single day with things like luxury goods, sex, certain chemicals, junk food and of course right there in the center technology all of these things activate your nucleus accumbens every single day now old and milner and much of the psychology community for decades believe that the purpose of the nucleus accumbens was to activate pleasure, right? Why else with lab animals and later people incessantly activate his heart of the brain if it wasn't because it felt good, right? Not exactly. It turns out what we now know that old and miller never did is that the way the brain stimulates us tow act is by creating this stress of desire this wanting reflect remember that exercise we did at the beginning of the workshop where I asked you to think about that message waiting for you blinking at you when I was doing was activating your nucleus accumbens and it turns out that that nucleus accumbens becomes most active in anticipation of the reward, but when we actually get the thing we want thing that we actually think is going to make us happy that's when the nucleus accumbens becomes less active so it turns out the way the brain gets us to act is by stimulating this itch that we seek to scratch this physiological response to a psychological stimulus write that message that you were thinking about you could see and that was causing anxiety was causing a bit of stress, curiosity, a physical reaction in many of you that that itch that we seek to scratch. And it turns out that there is a way to supercharge this desirous response, this stress of desire does anybody want to know how way I'm doing it to you right now? So when I took that long pause and I asked you a question with a question mark at the end, and I kind of changed my cadence and I stopped talking, some of you put up with this guy doing why did he stop talking what's going on? And it turns out that in all sorts of products, we find the unknown to be fascinating that a bit of mystery, a bit of the uncertainty causes us to increase focus, increase engagement and it's highly habit forming. This, of course, comes from the classic work of b f skinner. Many of you might remember sinner from your first psychology class back in college. Like, wanna one right? The father of operate conditioning skinner took these pigeons. He put them in a little box he gave them a lever to peck at. And every time they picked up this little disc they would receive a reward so at first pigeon would peck at the disk, get a little food pellet and skinner could train these pigeons to peck at the disk whenever they were hungry. But then skinner does something a little different. Skinner introduced a variable reward, so sometimes the pigeons would peck at the disk. Nothing would come out no food pellet but the next time the picture would peck at the disk, they would receive a reward and what skinner observed wass that the rate of response a number of times these pigeons pecked at the disk increased when the reward was given on a variable schedule of reinforcement. Why does this happen? Because we now know that variability spikes activity in the nucleus accumbens creating this desire iss response stimulating this itch that we seek to scratch, and so in all sorts of products that we find most habit forming, most engaging the things that capture our attention and won't let go. You will find one or more of these three types of variable rewards rewards of the tribe rewards of the hunt and rewards of the self letme introduce these to you rewards of the tribe are things that feel good that have this element of variability and come from other people. So the search for empathetic joy feeling good because someone else feels good competition, cooperation, partnerships all of these things feel good come from other people and have this element of mystery. This bit of variability best example I can think of online is social media, right? When I check my facebook account when I opened up that app, I'm never quite sure what I'm going to find, right? What photos might I see what the comments going to say? How many likes do something get high degree of social variability involved with a product like facebook? Stack overflow is another good example for any engineers who are it might be watching or in the room stack overflow is the world's largest technical question answer site five thousand questions get answered every single day and these air not easy answers these are things that take a lot of time for people to answer correctly. Why are they doing it? There's no money changed hands why are people devoting what amounts to a technical documentation? They're doing all this stuff for free? Well, what happens is anybody here you stack overflow? What happens when you post on answer to stack overflow? What happens to your answer? Will you get up ordered? Right of your tradition increases exactly what score you get up voted or down voted right and so there's this variability there's uncertainty about what my community, what my tribe off fellow engineers I might think about me it's not some cheesy gamification techniques it's not you know it's not some algorithm doing the voting about you it's really people it's my community, its people whose opinions I care about people in my tribe. Okay, so that's an example of rewards of the tribe rewards of the hunt rewards of the hunt are all about. The search for resource is right, and this comes from our primal search for food and other material possessions. And in modern society we buy these things with money. So when many people think of variable rewards they think about slot machines, they think about how you know the variable reward when you're playing a casino game is the uncertainty over on what might come out of these games of chance, we see very similar phenomenon with online behaviors like day trading or playing the stock market, right part of what makes day trading and playing the stock market even there's a lot of studies that show that day trading is a terrific way to lose money. Ah lot of people love this intoxicating game of the stock market going up in the stock market going down up and down, up and down we also see a similar psychology at work in retail right that shopping and in fact can have this reward otto hunte mechanic, for example ross dress for less right ross dress for less is not a company that most people think of as masters of persuasion but let me tell you they do an amazing job of people who liked of getting people who like to shop it ross to keep coming back and keep checking and checking the stores. Why? Because unlike its competitors j c penney and wal mart and target they stock the racks of ross very differently so when you go to a warm our tow her target or j c penny for example you go to your your section right you goto the men's or lady section and you see a particular blouse and you see that same exact item of clothing the same exact one in small medium large extra large right same exact thing but that's not how it works at ross at ross you go to your size and there you will find individual articles of clothing one different from the next to the next the next to keep you searching to keep you doing what they call the treasure hunt this searching and searching for the next deal so ross will actually ship individual items of clothing to individual store locations to put these little bits of treasure in their stores to keep people searching. In fact, when I talk teo many people who were raw stevo tease people who are habituated to shopping at ross what they told me was that they will go from ross to ross to ross in their neighborhood looking for these deals that is not something that happens in a wal mart right? Because you know what? You're gonna get a walmart say they're here to find the other walmart but the's ross's key people searching for these treasures these variable rewards of the hunt now how do we see this online? Turns out we see a very similar dynamic when? When? When we when we use our fees have you all noticed? By the way, how feeds aren't everything today? Why? What is every product online? Have this feed mechanic? What is it about the fee? Well, let's, take a look at lincoln's feet, for example. So when you open up the linked in ap all right, you start scrolling through so that's not very interesting in the second post is not that it very interesting, but all maybe the third is interesting. And what do I have to dio to get more of those rewards? What action is required? Scroll just that act of scrolling and that act of searching and scrolling and searching and scrolling is the exact same psychology that keeps us pulling on a slot machine searching and searching for that next reward. If you want to see the masters of this, take a look at pinterest another multi billion dollar company today, pinterest is the master of getting people to continue scroll, I dare you to go to the pinterest home page and not scroll once you can't do it, it's impossible, you'll have to scroll at least once, right? Because they designed this page to be enticing right to keep you scrolling. In fact, I I pulled this comic from a page that a pinterest user posted this on a page called pinterest addicts and it's got a little guy hunched over his computer and says, scroll, scroll, scroll, scroll, scroll, scroll! Then he says, hey and then scroll, scroll, scroll and girls, girls, girls that that scrolling and scroll, hoping to find the next reward. Searching and searching and never done searching for that next variable reward that's rewards of the hunt and finally, rewards of the self rewards in the self are things that feel good that have an element of variability but don't come from other people and aren't about these material or information rewards. Rewards of the self are things that are intrinsically pleasurable. They feel good in and of themselves. For example, the search for mastery, consistency, competency and control. Best example I can think of online is gameplay. All right, so when we play angry birds or dots or candy crush when we play, these online games were not necessarily winning anything. Now, there's no material rewards many times. We're not even playing with anybody else but there's something fun and engaging around getting some next level, the next accomplishment, the next achievement. Now some people say to me, yeah, okay, that's great near. But look, you know, I'm not much a gamer. This doesn't really apply to me, but I bet you play this game every day. This look familiar, right? Checking those un read messages in your in box clearing the two duis and your to do list or the one that always gets me is that one act that I've got on my home screen? I just have to open so I can clear it away are all examples of these variable rewards of the self, the search for mastery, consistency, competency and control? Now I want a few words of warning number one at that variable rewards are not a free pass variable rewards are not a free pass, so fundamentally there has to be a connection between the internal trigger and the variable reward right internal trees. We talked about them at length very, very important, and the job of the variable reward is to scratch that users itch. So if the internal trigger is boredom, well, then the variable reward has to entertain. Right and she kiss case the internal triggers boredom the variable reward has to entertain but if the internal triggers is seeking connection loneliness for example well then the variable reward has to connect people together so there has to be an authentic connection we have to scratch the users it many people oftentimes we'll ask meabout gamification and I'm not anti gamification right? Does anybody know what gamification is game using game like mechanics points badges leaderboards in non gaming environment is kind of got popular for a while then it's a little less popular but I think it actually has great applications where I see it falling flat is where it doesn't actually scratch the users it's where it doesn't give the person what they came for it's great for making something more entertaining when the internal triggers boredom really bad at alleviating fear or loneliness right? So we have to consider we have to make sure that there's a connection there that the that the variable reward is actually scratching the user's itch that's the first word of warning the second word of warning is that autonomy is a pre wreck quit it there's a psychological phenomenon called reactant and reactant tells us that people who feel like they're being controlled people who are feel like they're being told what to do rebel all of us do this so it's a natural human tendency so if you've ever had a boss that micromanage you if you've ever had your mother tell you to put on a coat when you're a perfectly capable adult right that feeling don't tell me what to do that's react it's and the problem here is where I see some companies get into trouble is particularly these products that try and do good for people and it's so fantastic help people save money or lose weight or exercise more whatever might be that they want to help people d'oh but then they start becoming these these acts that control us that tell us what to do so instead of telling users you know, some kind of reward these acts become variable degrees of punishment, right? They tell us hey, you're still broken, you're still fat so we have to be very careful that the experience itself the user is always in control it's something that the user maintained their sense of autonomy they're not being controlled, okay? And of course anything scam me or anything shady, you know people could detect that a mile away the user always has to feel they are in control. And finally the last bit of warning is to be careful of what I call a finite variability finite variability so something at some point in my presentation somebody will have the thought in their head that you know, I'm not sure if I buy this because look if variable rewards are so important and so powerful. Why aren't we all still playing the same games? Right? Think about super mario brothers or pac man, right? Like or far villas that remember farmville figures that it's actually pretty auspicious that were in this very building. Because this is where I think it used to be a right like right over there. S o you know why are we all still playing farmville member? This was the fastest growing game in history a few years ago. What happened? Who plays farmville today? Well, let's, think about this for a minute. Zynga, the company that built farmville. What was the next game? That zynga mate after farmville. Anybody remember? I think it was frontier ville. And after frontier bill came chef villain after chef bill came farmville too. And in the next village, the next billing the next bill and users started figure out, wait a minute, it's all the same game and what was once variable became predictable and the variable reward was no longer variable. So that's an experience that's an example of an experience that I call on infinitely there start a finite variable experience on experience where the the more the user engages with the product, the more predictable it becomes. And look, there are lots of businesses which are good businesses which produce products that have finite variability where you know what you're going to get for example very few of us we'll watch a movie in a movie theater walk out and buy another ticket to go see the exact same movie, right? We already know the happy ending we know how things are resolved when I can watch it again very few of us will get to the end of a book and say, oh, that was amazing! I'm going to read the whole book again there are some cold classic it happens from time to time, but then it's about, you know, the community of other star wars fans it's about the inner intricacies of shakespeare sonnets it's not the typical you know, when you know the ending and it's given away you know the spoiler you don't go re engage why? Because the uncertainty is gone there's no more variability, you know what's gonna happen. The news business is like this, right yesterday's newspaper is worth less than zero. We have to pay people to haul away those newspapers, right? Even though that information is still useful information but it's not new anymore. We know what happened already so it's not useful to us it has to these businesses are what's called studio models. They have to constantly create new new new new new, always creating new content these aren't bad business is you just need to be set up with the kind of business model that allows you to constantly crank out new experiences new content for example as opposed to products that have infinite variability these type of products as opposed to those have finite veiled variability. These products have higher degrees of variability built into the product there's more of an unknown associate with what users might find with products that have infinite variability. So for example, using a social network for example, linked and twitter facebook these products utilized information the content comes from our friends it comes from real people and real people go on vacation they post pictures they show us what their pugs look like they post articles they post other content there's a much higher degree of variability around what you might find when utilized these social products. Okay, the whole point of the variable reward face is to give users what they came for to scratch their itch and yet leave them wanting more having this bit of mystery associated with what they might find the next time they engaged with the product or so there's two types of of of how to deploy variability we can either insert variability into an experience as we saw many of those examples right products particularly that alleviate boredom thes type of products do that really well but then we can also for products that cater to areas of service where the product is inherently variable the situation is inherently available so for example google I would not want to insert variability into its search result that's silly but the core experience of searching right sending a search query by its very nature is variable right there's uncertainty where am I going to find the answer to my query so what google is doing is giving the user greater agency and control over something that's inherently variable uber is another great example somebody mentioned uber before you know uber part of what makes uber so fantastic is that it gives the user greater agency and control over something that they had very little control over in the past for example remember before uber if you call the cab company and you said okay, how long until the cab gets here you know when should I go to the corner? The answer was always the same ten minutes behind the corner ten minutes maybe the cab would be there in half an hour maybe forty five minutes you had no idea I'll be there be there in ten minutes we'll see there right? Well now here comes uber and with this interface they now are giving the user greater agency in control because I can see with his little pacman taxicab that's coming to get me I can see how far that taxicab is away and so that experience that's inherently variable is now mohr under my control, right? This itch of can I get two? Wearing it to go on time is satisfied better with this product, because now I have greater agency control over a variable experience.