Behavior & Responsible Game Mechanics & The Future of Gaming I Guess

Editor's Notes

• #2: - This title is kind of like my SXSW, I was all over the place\n- Education, healthcare, nonprofits\n- Went to SXSW interested in games and this hot "gamification" thing\n- Gamification, briefly, is the trend of using game mechanics for other purposes\n- Think the "points" system\n- If you don't know me, I loooove video games\n\n
• #3: - If you don't know me, I loooove video games\n- brief background:\n- My dad was a bomber pilot\n\n
• #4: \n
• #5: - My dad bought an Atari so he and the other bomber pilots could have something to do when they were on alert\n
• #6: - We brought it home when the period ended, and hooked it up to a BW TV, and I was hooked\n- that was 25 years ago\n\n
• #7: - big changes since then, but we seem to be at the edge of a big shift:\n\n
• #8: - Games are ubiquitous. 200 million people play casual games every month, about 75% of them women\n- very different market\n- Game models themselves changing:\n\n
• #9: - OLD MODEL: Get an extra heart\n - NEW MODEL: Acquire massive collections of stats\n\n
• #10: - OLD MODEL: Get a high score\n - NEW MODEL: Constantly accumulate points on a fixed identity\n\n
• #11: - OLD MODEL: Sell copies\n - NEW MODEL: Keep people playing as long as possible\n- Finally, games are tangled up with other activities. Gowalla badges, Twitter stats, Stackoverflow's rewards\n\n\n
• #12: - Gamification, a big topic at SXSW this year\n- But the most interesting ones I saw were more about a more abstract topic — human behavior patterns and how to change them\n- The more I though about these, the more linked they seemed to game systems\nSO\n- Behavior change is really difficult stuff, as we all know from trying to change our own\n- starts with a grasp of some of the barriers\n- So we're going to talk about some panels, some cognitive issues, and then bring it back around to games\n\n\n\n
• #13: - Let's talk about brain bugs\n\n
• #14: - You can ask people to do things, or change their ways, over and over\n- You can even get them to do it for a little bit\n\n
• #15: An example is getting sick, getting assigned medication to take ... taking it for a while ... then stopping, mid-run.\nInitial stimulus, behavior change, then drop-off\nDieting can be like this\n
• #16: Advertising is similar: really sticky ads and ideas can create short-term behavior change\nBut in the long-term, we find that ads like this smoking one don’t create real change\n\n- Anybody who's tried to diet or give up twitter or whatever knows that simply reminding yourself over and over doesn't cut it\n- Even putting up little notes doesn't help as much as you'd think\n- You get acclimated to the stimuli\n\nWhat are these systems missing?\n\n
• #17: - So it's not the change initiator that does the lifting, it's the feedback\n- True behavior change comes when you a) do it and b) get feedback\n- Systems that don't have are feedback very difficult to "make sticky"\n- Systems that have repetitive feedback also fail: \n- When I jog every day for a week, I bet you feel great! Every day!\n- Then you stop.\n- Even though I'm getting a "reward", it gets old\n\n
• #18: - Jennifer Maer, from IDEO, use design to reduce unplanned pregnancies\n\n
• #19: - If you don't know, they're a big deal\n- 5 billion annually, and they cost the women involved an average of $1600\n- Birth control is has TERRIBLE feedback system\n\n\n
• #20: - You do it right: nothing happens\n\n
• #21: - You do it wrong: something happens......eventually\n- And it’s kind of binary/permanent\n
• #22: - The site IDEO made, bedsider.org, has a really fun tone and a unique approach to choosing methods, breaking them up into activity-based categories\n- A loooooottttt of research and testing went into this, but the site is largely informational — you view it, you leave it, you never go back.\n- The main behavior they focused on was reminding women to use birth control on a daily basis\n\n
• #23: - And the big solution, was two-fold:\n - One, daily text messages reminding people. Good old old-fashioned text messages with a cute message and a reminder\n - And it worked — at first. \n\n - Text messages aren’t feedback, feedback would be getting a high-five when you take the pill\n - They’re just stimuli, which has diminishing returns\n
• #24: - The final step was to add enough randomness by joggling the times\n - And to custom-write each message, adding jokes, seasonal info, weird facts, etc\n- And they did it! Nearly 80% of participants reported that they used their systems much more regularly\n- The difficulty was the bad feedback system, and how to work around it\n\n
• #25: \n
• #26: - By the way, check out Pragmatic Learning and Thinking if you want to read some more bugs\n\n
• #27: - One that applies a lot in modern society is called "time discounting"\n- That's where we can't factor in the abstract, long-term effects of behavior\n\nNOT TO PICK ON SMOKERS\n- Smoking is a good example of this\n- Smoking has really bad feedback, but it's realllllly slowwwwww\n\n
• #28: - http://www.ted.com/talks/joachim_de_posada_says_don_t_eat_the_marshmallow_yet.html,,\n- Tangent: TED talk: Children who COULD overcome time discounting had a strong chance of future "success"\n\n- But we all know it's hard\n- It's really hard to save money\n- It's really hard to not eat that last piece of cake\n- To change behavior related to time discounting, you need speed up the feedback\n
• #29: - super-smart (ex-mozilla, wrote first geolocation spec) guy who's working for a firm that's trying to affect behavior in health\n\n
• #30: - Specifically, America's top diet-related diseases\n- Huge number of diseases on the killer list that are related to diabetes, heart disease\n\n
• #31: - Outlined some points:\n - Food feedback is slow\n - You can’t tell low-calorie from high-calorie foods\n - You can even eat so fast that you don’t feel full soon enough\n\n- As we said: Immediate feedback is REALLY effective\n- Providing feedback for abstract, difficult concepts is a good way to affect change\n\n
• #32: AS AN EXAMPLE OF FAST FEEDBACK\n- He talked about new cars (BMW?) that encourage efficiency by showing a constant MPG number\n - People drastically changed how they drove — coasting more, braking slower — just to drive this little number down\n\n
• #33: - And they started hitting stuff\n- But that's powerful!\n- How do we provide better feedback for eating?\n\n
• #34: - He told us some of their crazier brainstorm ideas\n- Aza said they'd been researching a in-ear microphone that could identify chewing AND food type\n- And speculated about magnetic dentures that make it harder to chew with repeated\n\n- ASIDE: The other idea was a toilet that measures your fatty contents and congratulates/criticizes you\n- "That's how you'll check in!"\n\n
• #35: - These are all examples of speeding up, or concretizing feedback\n- It’s like giving children a treat whenever they save money\n- Saving money is really slow and abstract, candy is good now, so you can build one with the other\n\n- By the same token, you can mess with the cycle to do evil\n\n
• #36: - This is why advertisers want to give you that immediate "reward" hit and bill you later\n- Buy now, pay later\n\n\n
• #37: \n
• #38: - Rebates (immediate feeling of savings)\n- Fast positive feedback, delaaaaayed and disconnected negative feedback\n
• #39: - So I've talked about making desired feedback fresh, and making it fast\n\n- The last bug is a little sneakier\n- You combine time discounting with this one and you can see how the brain gets in trouble\n- A trick for amplifying feedback, especially positive feedback\n\n\n
• #40: - News story from a few years ago: \n- Parkinsons is actually caused by the death of dopamine-containing cells in the movement center\n- Dopamine does a lot of things in several receptors\n- One of those things is distributing "rewards"\n\n- Certain actions/events you do are hooked up just right to the mesolimbic pathway, and you do them and you get a dopamine hit\n- The mesolimbic pathway helps you associate them with the the situation: the environment, your actions, etc\n- This pathway helps us look for patterns\n\n
• #41: - Parkinsons is often treated with an agonist, which simulates dopamine in its absense\n- ReQuip was a new agonist\n- Used for Parkinsons and some impulse control diseases\n- In some reports, nearly 20% of patients were experiencing this weird new symptom\n\n
• #42: - I mentioned earlier that the brain looks for repeatable reward patterns\n- “I do X and X happens, and it’s good”\n\n
• #43: - The problem is that our brain was built to do this in a difficult natural environment full of patterns\n- Finding food, discovering shortcuts, etc.\n- In a primitive environment, you have to learn very quickly that good things have patterns, and you have to find them to continue to get good things\n- What happens when the rewards have no pattern?\n\n- Back to ReQuip\n- The weird dopamine effect of ReQuip was\n
• #44: - An incredible gambling addiction\n- People with no gambling history just got hooked\n- One woman started gambling until 3:30 every night\n- Lost 200,000 dollars in a year\n- Doctor took her off the drug, never gambled again\n\n- This has led to a larger understanding of how rewards work, and how our brain is built to acquire them\n- And how things with random patterns - like slots and watching sports - have the weird effect they do\n- A current understanding is that random rewards “hack” the brain, and create a much stronger behavior link\n
• #45: - Why am I picking on game-like systems? After all, everything can be addictive\n- It takes a very long time to build, market, test a traditional product\n- And it's very hard to iterate on it\n\n\n
• #46: - Even on a regular community website... \n- It takes a long time to build slow rewards for users on a website — making them feel welcome, entertained, connected\n- This is metafilter’s top user, who I’m sure loves metafilter and gets a lot of positive feelings from it\n- Good for him\n- But it must have taken tooooons of work to get here, most of it thankless and lonely\n\n
• #47: - "Game mechanics" let you cheat a little bit\n- Is this as strong, as rich as metafilter? no, but it’s an example of creating “rewards” out of mid-air and distributing them\n\n- Knowing about these bugs: reward randomness and feedback speed, you see interesting corellations in games\n\n
• #48: - Call of Duty's random rewards (10,000!)\n\n
• #49: - Gowalla\n- StackOverflow's set of achievements, which tend to appear as surprises\n\n
• #50: - Farmville - random items and events\n\n
• #51: - Coindozer\n\n
• #52: - There are examples of this mechanic used for good\n\nSO: with these little behavior tricks,\n- You can make something sticky without necessarily making it meaningful\n- When we talk about gamification, are we talking about adding “play” to something?\n- Or is the point of a game system just to change behavior in our favor?\n- Or WORSE, is it just to make them play more?\n- This is on my mind because so far, "gamification" is picking up on the worst trend of games\n- Instead of “play”, games like these just encourage “grinding”, work for work’s sake\n- raise a generation of hard-working office drones\n- no thanks\n\n
• #53: GAMES WITHOUT CONCRETE REWARDS\n- games without purposefully work-like, repetitive structures\n- The other side of game systems is of course the "play" side\n\n
• #54: - Games like Scribblenauts, Minecraft, Little Big Planet, Dwarf Fortress, Canv.as, the Refrigerator magnet game\n\n
• #55: - 16bit computer\n\n
• #56: \n
• #57: \n
• #58: moot’s canv.as\n- They let people play with ideas, solve problems, and relax\n- They don't provide piles of achievement, self-worth, accumulation, identity\n- They don't work to "hook" gamers, and don't require constant interaction\n\n
• #59: - http://slidesha.re/hVLRgD\n- Sebastian Detering has a great slideshare on the gap between where gamification is now and where it should be\n\n
• #60: \n
• #61: And that’s all I have! Questions?\n