It is well known that a decision made by several people is often more successful than the judgment of a single expert.
Many modern institutions are based on this idea, such as elections, juries, financial forecasts, and even medical advisory boards. However, scientists have not reached a consensus on the nature of this phenomenon.
Some argued that to achieve the best outcome, it was sufficient to gather the answers that came to mind for the participants in the discussion and then draw an average conclusion. Others, on the contrary, believed that the best decision should be reached by consensus. The work of a team of Argentinian neuroscientists appears to have settled this debate, and their research was published in the journal Nature Communications.
A team led by neuroscientist Federico Barrera-Lemarchand from the University of Torquato di Tella in Argentina proposed an innovative hypothesis: that collective decision-making becomes more accurate when the group is broken down into simpler components, a method known as Fermi's evaluation. To test this, three experiments were conducted with 900 participants.
In the first experiment , participants were divided into groups of four. Each group was asked to solve several problems requiring numerical evaluation, such as "How many steps do you need to walk to climb a monument?" They coordinated their actions via text chat, without receiving detailed instructions. Researchers then analyzed the conversations and compared each group's answers. It turned out that the groups that spontaneously broke down the question into smaller parts—discussing the height of the steps, the number of steps, and so on—provided more accurate answers. It was also observed that these groups lacked prominent leaders; all members participated in the discussion.
In the second experiment , the researchers tried to determine whether participants could be trained in this tactic. Some groups were asked to work in the traditional way, that is, to exchange opinions and make a decision based on the average of the results. The other half were given clear instructions: first, to break down the problem into subtasks, evaluate each one individually, and then formulate a unified answer. In the end, the teams that received instructions to break down the tasks significantly outperformed the control group.
In the third phase, the effectiveness of Fermi's method was compared between individual and group work. All participants applied the same scheme to break down the problem into components, but individuals performed worse than groups, suggesting that group discussion and mutual hypothesis verification offer an added advantage.
The researchers paid close attention to the linguistic analysis of the participants' dialogues. It was found that the groups that successfully solved the problems used more concrete words directly related to the issue. For example, in the problem of evaluating the number of steps, the words used were "step," "floor," "height," and "step." In contrast, the less successful groups tended to use abstract terms. This demonstrates that problem-solving requires not only mental effort but also rich, meaningful communication.
Level
ReplyDelete