Discriminating the relevance of web search results with measures of pupil size

Abstract

The overwhelming amount of information on the web makes it critical for users to quickly and accurately evaluate the relevance of content. Here we tested whether pupil size can be used to discriminate the perceived relevance of web search results. Our findings revealed that measures of pupil size carry information that can be used to discriminate the relevance of text and image web search results, but the low signal-to-noise ratio poses challenges that need to be overcome when using this technique in naturalistic settings. Despite these challenges, our findings highlight the promise that pupillometry has as a technique that can be used to assess interest and relevance in web interaction in a non-intrusive and objective way.