I designed multiple visualizations to highlight the work in algorithmic behavioral science done through the Center for Applied Artificial Intelligence (CAAI) at the University of Chicago Booth School of Business.

Supervisor: Sendhil Mullainathan

Description: The scientific method typically starts with some sort of data, which leads to a hypothesis, then we do some science and end with some empirical tests that allow us to prove our hypothesis (or not). The data that we start with can be from introspection, anecdotes, observations, or memories - but all of these data have a human element to them. This human element can be limiting due to confirmation bias, sterotypes, bounded rationality, inattention, etc. so what if we used algorithms as our data to help us generate hypotheses?

I created three custom visualizations to catch the eye of the reader and to draw them in to different aspects of the research. In “My Face” the user uploads an image of their own face and the page returns the predicted value of their likelihood of winning an election. The “Face Game” gives users a chance to see if they can beat the algorithm at predicting the winner of elections, based on face alone. In “Face Space” the user can explore the dimensions of possible faces by moving sliders to see what the algorithm produces.

Repositories/Websites

Final version: The Face Effect: the published version of the project, completed in March 2023.

  • A tweet thread from the Center for Applied AI at Chicago Booth

CAAI-faces repository: the code

CAAI faces: my original work on the website from August 2021.