Tyler Cowen gets the best email. Case in point is this advice from a former cab driver on the best way to get your stolen car back:

If your car is ever stolen, your first calls should be to every cab company in the city. You offer a $50 reward to the driver who finds it AND a $50 reward to the dispatcher on duty when the car is found. The latter is to encourage dispatchers on shift to continually remind drivers of your stolen car. Of course you should call the police too but first things first. There are a lot more cabs than cops so cabbies will find it first -- and they're more frequently going in places cops typically don't go, like apartment and motel complex parking lots, back alleys etc. Lastly, once the car is found, a swarm of cabs will descend and surround it because cabbies, like anyone else, love excitement and want to catch bad guys.

What do waiters in a 1920s Venetian restaurant and today’s average role-playing game fan have in common? They both tend to remember what they have yet to finish.

Sometime during the 1920s, Russian psychologist Bluma Bulma Zeigarnik was sitting in an Austrian restaurant (or maybe German; accounts differ) when she noticed something peculiar: waiters displayed an unusual ability to remember complex orders while they were being filled, allowing them to deliver the right combination of food to the right tables. But oddly, that information vanished from memory as soon as the eats were put in place (or maybe it was after the bill was paid; again, accounts differ). It didn’t seem to have much to do with sustained mental effort or chanting the incomplete orders under their breath to hold them in short-term memory. Instead, the orders that hadn’t yet been filled just seemed to nag at the waiters’ minds until they were checked off as complete.

Back in her lab, Zeigarnik pursued this idea and ran some experiments (Zeigarnik, 1927) involving the completion of various tasks or puzzles. Some of the subjects performing the tasks were interrupted, then everyone was asked to describe what tasks they had done. Like the waiters remembering what orders still needed to go to what tables, subjects were far more likely to recall the tasks they had started but hadn’t completed.

This “Zeigarnik effect” subsequently entered the psychology lexicon to describe how we tend to find it easier to recall a task –and the details surrounding it– when we feel like we have begun to undertake it, but been unable to complete it. Apparently we as humans don’t like it when we begin something and don’t finish it, and such circumstances create an internal tension and preoccupation with the task. Completing the task provides closure, release of the tension, and –not to put too technical a term on it– goodie feelie type feels.

When I played Skyrim, this is all I could picture at night when trying to go to sleep.

And we see this all the time in games, particularly role-playing games where in-game journals get crammed full of unfinished quests, errands, and tasks. Ever wring your hands over a huge list of incomplete quests or feel hesitant to progress the main quest until all those little side missions are checked off? To see the effect in action the next time you play a RPG try at the end of your gaming session to recall as many open quests as you can relative to completed quests.

And MMOs like World of Warcraft are the worst with this stuff, as anyone who has collected 13 out of 14 Goretusk livers can attest. Once we begin one of these tasks, they hang around in the back of our mind and are much easier to recall than completed tasks. Researchers Scott Rigby and Richard Ryan specifically call out the Zeigarnik effect in their book Glued to Games: How Video Games Draw Us In and Hold Us Spellbound:

MMOs are designed so that your list of tasks is never done. No sooner do you complete one, then two more pop up to take its place. Like a digital game of whack-a-mole, accomplishment only brings more unfinished business. As soon as you finish or “turn in” a quest, you are immediately offered another one with an even bigger reward. Or perhaps completion of one “unlocks” the opportunity to receive many new quests at once. Quests are often linked together in a series that helps move a story along, but never provides much closure. If I need to find 12 jewels to complete my quest, I will not stop at 11.

We also see this happen in empire building game series like Sim City or Civilization. The “just one more turn” is a direct result of the Zeigarnik effect, since that one more turn is almost always in the service of completing some structure, upgrade, technology, or conquest.

Familiar with the “just one more turn” phenomenon? Isn’t it funny how it usually happens because you’re waiting for some upgrade to finish?

And researchers have continued to study the Zeigarnik effect and refine models associated with it. Schiffman and Greist-Bousquet (1992), for example, found that people over-estimated how long they spent on interrupted tasks, even when the time spent was the same as completed tasks (probably compounded by the availability heuristic, which causes us to overestimate how big or frequent something is based on how easy it is to recall examples of it). Other researchers have noted how the Zeigarnik effect isn’t completely reliable and have explored moderating factors like how motivated people are to do the task (e.g., Reeve, Cole, & Olson 1986), the nature of the interruption, or task difficulty.

But regardless of whether the effect shows up every time a task is interrupted, it does often happen. The next time you find yourself thinking “just two more turns until this research is complete” or “I just need to kill the trolls to make this quest log entry go away” remember Bulga Zeigarnik and her waiters filling orders for hot dogs and waffles. Or something. I honestly don’t know what they eat in Austria.

Reeve, J., Cole, S., and Olson, B. (1986). The Zeigarnik Effect and Intrinsic Motivation: Are They The Same? Motivation and Emototion, 10(3), 233-245.

Rigby, S. and Ryan, R. (2011). Glued to Games: How Video Games Draw Us In and Hold Us Spellbound. Santa Barbara, California: Praeger.

Schiffman, N., Greist-Bousquet, S. (1992). The effect of task interruption and closure on perceived duration. Bulletin of the Psychometiric Society, 30 (1), 9-11.

Zeigarnik, B. (1927). Uber das Behalten yon erledigten und underledigten Handlungen. Psychologische Forschung, 9, 1-85.