Intertemporal Prospect Theory Paper (PDF)
Immanuel Lampe and Matthias Weber

Note: We received the Vernon Smith Young Talent Award for this paper from the Society for Experimental Finance.

Prospect Theory is the most prominent contender of expected utility theory to describe decisions under risk. In atemporal contexts, prospect theory is well understood. In intertemporal contexts, however, it is not clear how prospect theory should be applied (in particular, whether probabilities should be weighted within time periods or whether the probabilities of present values should be weighted). It is also unclear what parametric specifications of probability-weighting and value functions should be used. We find in a pre-registered experiment on a representative sample that an application of prospect theory weighting probabilities of present values predicts decisions best. Estimated probability weighting functions are very similar to those typically estimated in atemporal settings, while value functions are almost linear with a loss aversion coefficient close to one.

Civilian Evacuation During War: Evidence from Ukraine Link to paper
Seung-Keun Martinez, Monika Pompeo, Roman Sheremeta, Volodymyr Vakhitov, Matthias Weber, and Nataliia Zaika

In the media: Vox Ukraine, HSG Newsroom (German version).

In times of war, evacuating civilians from conflict zones is of critical importance for their survival and well-being. However, many people are hesitant to evacuate. Text-based nudges are a promising, yet unexplored, tool to increase willingness to evacuate. We conduct a controlled survey experiment in Ukraine in July 2022 during the ongoing war against Russia, varying the framing of, and information provided in, automated alert messages. Our findings suggest that providing individuals with an evacuation plan in the message is crucial. The specific framing of the message itself does not play a significant role in the perceived effectiveness of the messages. Observational data on actual evacuation behavior from the same sample of individuals supports this conclusion–finding that those who were offered evacuation assistance from people outside of their household were more likely to flee to safer areas.

Fake News in Social Networks Link to paper
Christoph Aymanns, Jakob Foerster, Co-Pierre Georg, and Matthias Weber

We propose multi-agent reinforcement learning as a new method for modeling fake news in social networks. This method allows us to model human behavior in social networks both in unaccustomed populations and in populations that have adapted to the presence of fake news. In particular the latter is challenging for existing methods. We find that a fake-news attack is more effective if it targets highly connected people and people with weaker private information. Attacks are more effective when the disinformation is spread across several agents than when the disinformation is concentrated with more intensity on fewer agents. Furthermore, fake news spread less well in balanced networks than in clustered networks. We test a part of these findings in a human-subject experiment. The experimental evidence provides support for the predictions from the model. This suggests that our model is suitable to analyze the spread of fake news in social networks.