I'm an Assistant Professor of Information and Computer Science at the University of Michigan.
My research explores people's user interface preferences and how they are influenced by culture. I'm interested in teaching computers to behave "culturally intelligent", so that they learn to automatically adapt themselves to what we like and what we can work with most efficiently.
To do this, I collect large amounts of data from people around the world with our experimental platform, the LabintheWild. LabintheWild is a virtual lab for conducting behavioral studies that lets participants compare themselves to others.
My interest in various cultures has also had the side effect that I frequently have to get away from the computer and actually talk to people. In the real world. In different (sometimes scarily different) countries. I got my first cultural shock when working with farmers and people from the agricultural ministry in Rwanda, but very quickly fell in love with the country and its people. I went back in 2009 to teach a course at the National University of Rwanda, and to do some more research. In between, I got frequent cultural shocks when living in Switzerland (but quickly learned that they can be overcome with lots of chocolate). In Switzerland I also did user studies with many people from all over the world. You can find some of my observations in my publications. If you are interested in even more details, here's my CV.
LabintheWild is our virtual lab for conducting behavioral research studies online. LabintheWild participants receive personalized feedback at the end of each study, which lets them compare themselves and their performance to people of other countries. In less than two years after launching the platform, LabintheWild was visited more than 2 million times, and nearly 750,000 visitors completed an experiment (an average of about 1,000 participants per day). Because our participants come from pretty much every country around the world, we are currently in the process of translating the platform and our studies.
Our preferences differ in what we like to see, what we find trustworthy, and appealing, and what we can work with most efficiently. Much of this is determined by our cultural background. The goal of this project is to develop a method for today's one-size-fits-all websites to adapt themselves to what we like. To achieve this, I use LabintheWild to collect large-scale data sets and develop models of people's preferences with the help of computational image metrics. I have also developed a culturally adaptive web application, which is able to compose personalized user interfaces based on the information about a person's current and former countries of residences. Feel free to have a look at my publications or to contact me if you are interested in hearing about future directions for this project.
Who would have thought that countries with a faster pace of life schedule their meetings much further in advance? Together with Doodle, the popular online event scheduling company, we are exploring cultural differences in the group decision-making process when scheduling events online. Among other findings, our CSCW paper from 2013 shows that people from collectivist countries (e.g., India, China) make a larger effort to reach mutual agreement. We are currently working on a qualitative follow-up to this work, and on trying to predict people's answers to Doodle polls.