The Power of Anonymity: A Conference Experiment

Picture this: a bunch of scientists huddled over their computers, typing away at the UAI 2022 conference. The hot topic? Should reviewers know each other's names while discussing papers? The conference decided to test this out. Reviewers and papers were split into two groups – one where names were visible, and one where they weren't. Both groups discussed papers on a typed forum, and afterwards, reviewers shared their thoughts in a survey. The results? The anonymous group had a bit more chatter, less sway from seniority, and reviewers felt they had a better time. It's like when you're at a party and wearing a mask – you might open up more! This little experiment sheds light on how knowing someone's name can influence our behavior. In the group where names were shown, there was less discussion. Maybe because people were a bit more cautious or concerned about what others might think. And there was more influence from seniority, which makes sense – when you know someone's status, you might give their opinions more weight. But why did the anonymous group feel they had a better experience? Maybe because they felt freer to express their thoughts without worry. It's like being in a costume – you can be whoever you want to be, and say what you really think. This also ties into something called the "online disinhibition effect, " where people feel more comfortable sharing online when they're anonymous. So, should reviewers know each other's names? This experiment suggests that anonymity can lead to more open discussions and less bias. But it's not a one-size-fits-all answer. Sometimes, knowing names can help build community and trust. It's all about finding the right balance.