• cultural markets
• social cues
• heuristics
• decision making
• experimental economics

On the Internet, people have free direct access to information related to the popularity and the quality of products. In many cases, cues hinting to popularity and quality are made prevalent in the interface of the website in order to facilitate the choices made by the users. In addition, many websites use algorithms based on those two attributes to generate the order in which the users see the items. Although this information is an inherent part of many websites little is known about how people integrate these two sources of information.

In this study we examine how people use popularity and quality cues for books, films and homemade videos (available at www.amazon.com, www.imdb.com and www.youtube.com, respectively) to choose between alternative options. We implement a binary choice task in which items are presented to the participants simultaneously alongside with cues referring to popularity and quality. All other information (e.g., name, content, author) is withheld to focus on the interaction of these two cues. For each of the three experiments, the experimental task comprised of 200 choices, which were consequential for the participants.

We compare the predictive performance of heuristic, probabilistic choice and machine learning models over the collected data. Our results indicate variability between participants but consistent decision rules within each subject. Furthermore, we find that the valence of popularity and quality cues varies across the three websites, indicating environmental features that could lead to the emergence of different decision processes across them.