What is it about?

Our experiment shows that search engines can lead users to unknowingly exhibit cognitive biases. Even when reading only valid scientific information, four cognitive biases consistently emerged. This is problematic because it limits science discourse.

Featured Image

Why is it important?

We found search engines limit both scientific context and what people can do to find out more context. Due to the search engine's limitations, users would exhibit four different cognitive biases. For example, the information provided at the top of a search engine page is often presented in an "Answer Box." Users will tend to trust this answer more than other results without questioning the context of the scientific information. This is problematic because understanding a scientific topic often requires more context than a single source can provide. For example, one group of scientists may find solar power to be the most promising alternative fuel for a city to invest in while a second group of scientists advocates for wind power. However, if only one of these groups is presented at the top of the search page, then users may limit their understanding to the information provided by one group of scientists. This is just one brief example of the four different types of cognitive biases found in the paper. This study is unique because it found cognitive biases emerge in search engine pages presenting scientific topics with genuinely scientific sources (i.e., not biased sources or scientific misinformation). In addition, while it is commonly thought that the order of search results is the cause of these biases, this study found other factors play a role. Cognitive biases also depend on the type of search results presented. For example, a person who only looks for Wikipedia articles will have a different bias than a person who only looks for scholarly genres.

Perspectives

This study was conducted just as the terms "fake news" and "biased algorithms" were becoming more popularized in mainstream media. This study builds from work by Eric Meyers that found that children were starting to place too much trust in search engines for information and went on to inspire the dissertation for Alamir Novin on how cognitive biases emerge.

PhD Candidate Alamir Novin
University of British Columbia

Read the Original

This page is a summary of: Making Sense of Conflicting Science Information, March 2017, ACM (Association for Computing Machinery),
DOI: 10.1145/3020165.3020185.
You can read the full text:

Read

Contributors

The following have contributed to this page