Ordering at a restaurant may seem like a simple decision, but for many people it creates a dilemma: should you choose the dish you know you like, or try something new that might be even better?
Now, scientists say they have solved a decades-old puzzle, using mathematical modeling and large-scale experiments on human behavior. The study addresses the so-called “explore or exploit” problem, that is, whether to keep trying new possibilities or stick with your preferred choice.
According to researchers, the best strategy depends on how often you think you'll return to the same restaurant. If you're just starting out and expect to eat there many times in the future, it's worth trying new dishes, because you might discover something even better than your favorite.
But if you have few opportunities to return, then it makes more sense to choose the dish you already know you like. In other words, the closer your visits to that restaurant get to the end, the more you should trust the safe choice.
This problem originated about 40 years ago, when the famous physicist Richard Feynman was having lunch with his friend Ralph Leighton at a Thai restaurant in Glendale, California. Leighton was debating whether to order his favorite dish, ginger chicken, or try something else. Feynman turned this dilemma into a mathematical problem and solved it, but he never published the analysis.
For decades, only his handwritten notes remained of that conversation, which Leighton preserved. Researchers from Princeton University were later able to decipher them and reconstruct Feynman's problem and solution.
The mathematical model shows that at the beginning of a series of visits, it is more worthwhile to explore new options, because you have time to profit if you find a better dish. As the number of remaining visits decreases, it becomes increasingly reasonable to stop at the best dish you have tried so far.
To test this idea, the researchers involved 2,520 participants in decision-making tasks that mimicked the restaurant dilemma. The experiments varied according to the number of choices remaining, the quality of the best known option, and the uncertainty about the untried options.
The results showed that people naturally follow a similar pattern: they try more new things at first, and then gradually move to their preferred choice. In fact, participants tend to explore a little more than the mathematically optimal strategy suggests, especially in the early stages.
The conclusion of the study is not that you should always try something new, nor that you should always get the same dish. The best decision depends on how many times you think you will have to eat at that restaurant or in that city again.
So, if you're in a place you'll be returning to often, try something new. But if it's probably your last visit, the wisest choice might be the dish you already know won't disappoint. /GazetaExpress/
t7 live
