People correctly define the strategy of adopting successful solutions even in the case when the task does not involve this, and the outcome is purely probabilistic, but not always use it. It found American scientists: they conducted an experiment in which participants had to select a winning option based on a certain pattern, either unknown to them the probability of an outcome. In fact, and in another case the participants were able to learn a successful combination, basing its strategy on the changes in the environment, but a probabilistic event still it is not used as it did not provide a big win and was, in fact, useless, write scientists in Nature Communications.
Any decision that a person takes in a particular environment due to two factors. The first type of factors is called a descending (top-down) — they include all known in advance information and the second upward (bottom-up) based on additional data, which may affect the decision in the moment. In fact, decision-making involves the confrontation of these factors: it is possible to rely on already known information that will provide the outcome with a certain probability, and this probability can predict given new data.
It is not always clear which option is best. On the one hand, to ignore the new information in the environment may in fact be inefficient, but on the other — already well-known and many times the path to the solution may be the only true. To study decision making in such a situation more decided Konovalov, Arkady (Arkady Konovalov) and Ian Krajbich (Ian Krajbich) from the Ohio state University. For this they used an experimental paradigm in which the rate of resolution of the participant and his confidence in it is the movement and the position of the mouse.
The participants (there were 57) on the screen showed a picture of two fractals, which I had to choose one. After that, different probability appeared another image, but not immediately: it was necessary to move the cursor to the bottom of the screen, and then click the mouse on the image. Then appeared the winning amount (determine the final monetary reward, which will be provided to participants). The probability of occurrence of a fractal has been defined in advance, but for the participants was unknown and random, so no pattern, which could be learn in order to get the maximum reward, it was not.