Somehow, working together generally produces better results than working alone. Ever wonder why? Well, researchers from Carnegie Mellon University have created a model that explains how collective group decisions are made, even when all members do not have access to all available information, and neither can they communicate to make a final decision. The ‘de-centralized decision-making’ demonstrates how positive feedback during the investigation process assists in making effective and efficient decisions.
Developing Statistical Model: Decision-Making And Accuracy
During the primary selection process of the presidential campaign, everyone is trying to find an ideal candidate in a rather crowded space. The person in the lead tends to get more media attention and coverage. This might lead people to think about voting for him, based on all the hype and recognition and continuous exposure.
David Hagmann, a Ph.D. student in CMU’s Department of Social and Decision Sciences explained an alternative scenario, commenting that sometimes the added media coverage highlights rather notorious information about the candidate, ultimately failing to gather votes and fading in the polls.
Hagmann, along with Russell Golman and John H. Miller, developed a mathematical model that featured two elements. The first is enrollment with positive feedback – initial reinforcement of popular choices. The second is quorum sensing – sufficient support from peers regarding the initial choice triggers the final decision.
Using a statistical model known as a ‘Polya urn scheme’ (balls of various colors are drawn repeatedly from a container and previously picked ones are more likely to be selected again), the research team was able to observe how long it took to actually make a decision, and to calculate how accurate that decision was.
Model Provides Insights Into Human Brain
According to Golman, Assistant Professor of Social and Decision Sciences, the model was found to be extremely robust in its implementation. An interesting aspect: a choice that has a lot of variation in its perception is chosen less frequently, thus establishing a systemic aversion of risk.
“When everyone has to do the same thing, you want to be slow and steady to avoid extreme choices”, Golman explained.
This model could provide valuable insights into the brain’s neural networking. For the latter to function effectively, a certain number of neurons need to be activated to make a decision. Unlike the model, currently existing theories regarding neuronal decision-making do not consider the aspect of positive feedback. Golman also stated that neuroscience generally acknowledges that neurons are connected in recurring networks, which makes way for positive feedback and reinforcement.
Moreover, the model also explains how famous trends take off, such as the popularity of a gadget, and the success behind ‘word-of-mouth’ marketing schemes.
“Choosing the most popular gadget is not necessarily the best decision, but it’s not a bad rule of thumb”.
The interesting study appears in Science Advances.