When do humans and bees act like "it won't happen to me?"

This is according to an article published yesterday in Nature by researchers from the Technion, Tel Aviv University and the Faculty of Agriculture of the Hebrew University in Rehovot in an unusual combination of psychological research in humans and observation of animal behavior

Experimental research on decision-making among humans reveals a considerable difference between decisions based on description and decisions based on experience. When people rely on a description of low risk, the issue of safety is of great concern to them; On the other hand, when they rely on personal experience, they act like "it won't happen to me." Similar results were obtained in bees. A more careful analysis of the effect of experience shows that accurate feedback increases the tendency to ignore undesirable outcomes that are unlikely to materialize.

The study, published yesterday in the prestigious scientific journal Nature, shows that people are more inclined to stick to certainty when the difference between potential outcomes - for example, winnings in a gambling game or returns on financial investments - is difficult to estimate. The research was led by professors Ido Arab from the Faculty of Industrial Engineering and Management at the Technion, Arnon Lotem from Tel Aviv University and Sharoni Shapir from the Faculty of Agriculture of the Hebrew University in Rehovot. The three gave the volunteers a choice between two buttons on the computer - R (dangerous) and S (safe). Pressing S led to a reward of 3 points with full certainty, and R led to a reward of 4 points with certainty of 0.8; It should be noted that the participants learned about this mechanism by trial and error. Participants tended to take a risk strategy when these results were presented in simple numbers. However, when the numbers 3 and 4 were converted into clouds of 30 or 40 points, effectively blurring the obvious differences, the participants changed tactics and now tended to the more certain results.

The researchers conducted a similar experiment in bees - where sugar solutions in different concentrations were used. The results were similar, in other words - we are no different from bees in the way we consider risks. It turns out that humans and bees basically use the same decision-making mechanism, and make the same errors, when they have to choose between receiving a certain reward and receiving a higher reward but with a lower probability.

The inspiration for the research comes from a well-known phenomenon called the certainty effect, according to which people would prefer to win a reward of 3 with certainty over 4 with 80% probability, but would not prefer to win 3 with 25% explanations over 4 with 20% probability (despite a ratio the same between the probabilities). That is, it seems that a sure win is particularly attractive and that humans are generally risk averse.

Later experiments with rats suggested that animals also show the same tendency, but more detailed studies done with humans showed that when humans have to make a decision based on an examination of the two alternatives by way of trial and error, without the chances of reward being explicitly stated, they show a tendency to prefer risk . That is, when they are actually tested like the rats, and they have to alternately press buttons, they will prefer to press the button that gives 4 with 80% probability and not the one that gives 3 for sure. The new study by the Israeli team provides a general solution to the problem. He shows that when humans and animals have to make a decision based on trial and error, they explore both options and develop a preference for the alternative that is pictured in their memory as being better most of the time. It is important to remember that using this simple rule does not always give the mathematically preferable result. Since this rule does not take into account the size of the differences in reward but only who seems to be greater, it is particularly sensitive to the ability to distinguish between the different rewards. A clear view of the rewards helps to distinguish who is usually greater, therefore encouraging risk taking. A less clear display makes the magnitude of the reward of the safe option seem equal to that of the unsafe one, hence the preference for the certain reward. The researchers showed that by changing the experimental conditions in a way that affected the ability to distinguish between the different rewards, both humans and animals can be made to prefer risk or certainty even when the rewards themselves have not changed.

The research has possible implications for understanding decision-making processes in everyday life, and for the evolutionary background to human behavior.

How they arrived at the comparison between bees and humans

Sharoni Shapir says in response to the science site's question how he came up with the idea of ​​comparing bees and humans in the first place: "I have been dealing with the decision-making process of bees for several years. For several years we have noticed that bees are a good model for decision-making processes in other animals as well as in humans."

"What is interesting is that it is also a good model for the irrationality of decision-making. The fact that people make irrational decisions sometimes in a consistent way is known from experiments with people that we do not make decisions according to the classical theory in economics and Daniel Kahneman even received the Nobel Prize for this. But in animals like It is very important for a bee to make the right decisions because it involves its life and it has millions of years of evolution for them to do it right. What we see is that many patterns of decision making are also true for bees and not only for humans.

The same apparent psychological effects work on bees as they do on people, so this also means that plants can use the same marketing strategies that economics uses to attract bees.

The last experiment was another very interesting example because the certainty effect is a very strong effect in people and there are thousands of citations of Kahneman's classic articles. Even people like bees who have to experiment with the decision and do not have an exact description of it get exactly the opposite effect.

And why is there a difference in choosing between 30 and 40 and choosing between 3 and 4?

"You see the number 3 and 4 with your eye, obviously 4 is better than 3. If you go for 4 you will get more than 3 because 80% of the times you will get 4 and in total you will get more (bet effect), when it is 40 or 30 points on the screen is hard to tell them apart and then you play it safe (confession effect) so in such situations you have to experience it, so your ability to distinguish between the quantities will determine whether you play it safe. This is equally true in humans, bees and probably Even in previously tested rats."

It's not easy to submit an article to Nature, how did you do it?

In my opinion, the topic touched them thanks to the fact that we showed that a very simple and basic model can explain the behavior of both humans and bees, what's more, it explains a very important effect in economics and psychology, which is why Nature agreed to publish it and even promoted it in a popular article that explains it, something they don't do with any study.