People are influenced by unreliable information yet remain aware of its influence on their decisions

In today's digital landscape, individuals rely on a myriad of sources for decision-making, including social media, news, online reviews, and personal interactions. To assist in discerning the credibility of these sources, numerical indicators and labels have become commonplace, such as user ratings and fact-checking systems. However, an intriguing question arises: when these indicators are made explicit, do people utilize them effectively when making choices?

To investigate this matter, researchers from the University of Tsukuba and Queen Mary University of London carried out a series of experiments. Participants engaged in a decision-making task, which required them to consider information from multiple sources, each accompanied by a label indicating the likelihood that the source would provide accurate information.

The findings revealed that participants tended to give more weight to sources with higher reliability. Nevertheless, they did not employ the stated reliability values precisely in their decision-making process. Perhaps most striking was the observation that information labeled as having a 50% probability of being correct, which ought to provide no useful basis for making decisions, still significantly influenced participants' choices.

In a subsequent experiment, participants were also asked to report the extent to which they believed each source had impacted their judgment. The results indicated that although participants were able to recognize, to some degree, when unreliable information had swayed their decisions, they still allowed the influence of such information to guide their choices.

These findings, published in PLOS Computational Biology, underscore the importance of providing clear information about reliability and understanding how people interpret and incorporate this information into their decision-making processes. The results may offer valuable insights into how individuals respond to information of varying trustworthiness in the modern information environment.