Last week our 3-year-old daughter Zoe moved from her crib to a "big girl bed." Some kids make this transition smoothly, but not Zoe. After discovering she could climb out of bed any time she liked, Zoe decided to fully exploit her newfound freedom. And so for six consecutive nights, Zoe exercized that freedom at all hours, and we became sleep-deprived zombies. Through sheer desperation, we also became game designers.
We attempted to gamify the transition from crib to bed, but we proved to be inept designers. We saw the transition from crib to bed as an end-goal - a victory in itself. But it was our victory, not hers. Zoe needed a stickier, more tangible outcome, and imparting simple praise ("We're proud of you!") wasn't enough to keep her dialed in. Zoe needed a progression system that made sense to her, with meaningful feedback and a win condition. We lost a lot of sleep before we figured that out.
Zoe's bedtime game started promisingly. We began talking about a new bed two weeks before Christmas. We perused catalogs with pictures of children's beds; we discussed how all of her older cousins and friends sleep in beds; and we read books about bedtime. By the time we arrived at the furniture store to make our purchase, Zoe was pumped. We walked in the door, and Zoe immediately approached the sales clerk and said "I want a big girl bed!"
Downloading the demo
We looked at bed after bed, and Zoe laid on each one, pretending to sleep. Soon, however, she lost interest in shopping and discovered a more fun activity: hiding behind sofas and recliners, challenging us to find her. We refocused her attention on the task at hand (with threats of leaving the store bed-less), and she finally narrowed the field to two kid-sized beds. She tested each multiple times before finally settling on her choice. We bought it and arranged delivery in three days.
While we waited for the new bed to arrive, we let Zoe decide where in her room it would go. She chose a quilt from her grandmother to hang above her new bed; she chose a new blanket and pillow; and she helped us shop for a new night-light. Zoe was ready for her new bed. Better yet, she was invested.
Upgrading the hardware
On the day her new bed was to arrive, Zoe watched me dismantle her crib. This was a far more emotional event for me than for her. As I removed each bolt, I wistfully recalled assembling it before she was born. Zoe's concerns were purely practical. "Where should we put it?" she asked. "It's going to the garage for now," I told her. "That's okay," she replied, "It can get snow on it because I don't need it anymore."
Removing the shrink-wrap
Zoe was at pre-school when the bed arrived, so we had time to get everything ready. When she returned home, she ran upstairs, dashed to her room, and gasped "My new bed!!" Frenzied bouncing and running around ensued. She was thrilled. We were thrilled. The new bed was here. All was well.
That night Zoe eagerly climbed into her new bed. We read stories and tucked her in. "Do you like your new bed?" I asked. "Yes!" she exclaimed. "Go to sleep now, and we'll see you in the morning." "I will. Night-night, mommy. Night-night, daddy" "Night-night, sweetheart."
We turned off the light, closed the door, and gave each other a silent thumbs-up. Another parenting milestone passed. ... Thirty seconds later, the door opened and Zoe emerged. "Mooooommmy." This game had only begun.
Somewhere around the 15th time Zoe came out of her room that night, it struck me that we had identified the wrong challenge. We thought the bedtime game was about convincing Zoe to embrace change and accept her new bed. We were wrong. Introducing the bed was an easy test. The challenge was keeping her in it. We had no game plan for this.
Over the next several nights, Zoe ran us ragged, repeatedly emerging from her room and forcing us to put her back in bed. On the second night, in the space of less than an hour, I returned Zoe to bed 23 times.
Every resource we consulted advised a simple trategy: do not let the child win. Return the child to bed, no matter how many times it takes, until she finally falls asleep. Stay on this course until the child eventually learns that bedtime means sleeptime.
After five nights of heeding this advice with no progress whatsoever, we decided to devise an incentive. We drew a chart with pictures of her favorite things - ice cream, an iPhone game, a photo of her teenage cousin, etc. - spaced over time. We explained to Zoe that if she could stay in bed for one night, she would be rewarded with a trip to Dairy Queen. More consecutive nights would bring better rewards, culminating in her cousin (whom Zoe worships) arriving at bedtime to read her stories and personally putting her to bed.
This incentive system worked like a charm, and the last thing Zoe said to me as I left her room that night was "I get ice cream tomorrow." "Yes," I replied, "if you stay in bed." "I will," she responded. And she did. (Well, she did get up twice for a drink of water - a familar canard we chose to overlook). We went to bed happy that night.
Five hours later at 1:00am, Zoe walked into our bedroom. "I don't like my bed." We calmly returned her to her room, tucked her in, and reassured Zoe that hers was the best of all possible beds. The next several hours were spent replaying that sequence until she finally fell asleep for good.
The next night was more of the same. We took turns returning Zoe to bed. We tried the soft approach; we tried the hard approach, but nothing worked. Neither of us us were sleeping, and we were losing our minds. Some parents lock their kids' doors. Some use gates. Neither of those options appealed to us.
Help us, Jenova
In a state of desperation, my zombie-mind flashed on an idea - or, rather, an idea-man. For some crazy reason I thought of Jenova Chen.
If you want to teach players not to do something, you don't need to smack them. You need to give them zero feedback.
Negative feedback, Chen explained in a talk I attended last summer, is no less stimulating than positive feedback. Players often choose to "behave badly" in games simply to experience the resulting outcomes. If a designer wishes to dissuade players from choosing certain actions, he must associate no outcomes to those actions.
When Zoe walked into our room at 1:30am, was she essentially testing the system we had designed? In that system, she could interrupt our sleep at any time and receive a stimulating response, harsh or reassuring. What if we gave her nothing but dead air? What if, the next time she entered our bedroom after midnight, we simply ignored her?
And so the next night we went to bed and awaited Zoe's inevitable arrival. Right on cue, she popped in at 1:15am with "I don't want to sleep in my bed." We ignored her, and she repeated the line about ten more times. We remained still and silent.
Zoe seemed stunned. She walked to both sides of the bed as if to examine us. "Daaaaaaaddy.... Mooooommy." Dead air. She paused for a moment as if to consider her options...then she turned, walked back to her room, closed the door, and slept peacefully until morning. In her new bed. Like a big girl.
Thanks for the shut-eye, Jenova. We owe you one.