Researchers from the University of Toulouse-CNRS in France and the University of Leeds in the UK have explored how groups of experts make decisions and whether disagreement among them affects the reliability of their choices. Their study, published in the Journal of the Royal Society Interface, challenges the common assumption that high levels of agreement or unanimous decisions are always a sign of a group's expertise or trustworthiness. Instead, the findings suggest that when group members are overly confident in their decision, it can sometimes reflect a psychological tendency known as conformity bias — the tendency to align with the group even when it might not be the best choice. To investigate this, the researchers developed four different mathematical models to represent how conformity affects decision-making. They varied the complexity of the decisions being made, creating 16 different scenarios. For each scenario, they calculated the likelihood that the majority of decision-makers were correct based on how much they agreed with each other. This helped them assess whether high levels of group confidence were a good indicator of the quality and reliability of their decisions. The results showed that when conformity bias was not present, the accuracy and reliability of group decisions increased as agreement among members grew. However, when conformity bias was a factor, high levels of agreement — such as unanimous decisions — were no longer reliable indicators of trustworthiness or decision quality. The study also found that in more complex decisions, some level of disagreement could actually be a sign that group members were less influenced by each other, which might lead to more accurate outcomes. The researchers believe their work could inspire further studies on how cognitive biases influence decision-making, using mathematical models to better understand group behavior. Their findings could also help improve existing models of collective decision-making, potentially leading to new strategies for enhancing the reliability of group decisions in fields such as politics, business, and science.