Available here 

• Nonlinear Pricing Under Inequity Aversion | with Murat Yılmaz | 2020, Journal of Economic Behavior & Organization | [link] 

• Active Participation Bias | with Thomas Dohmen |  [latest version available here]
  Abstract: The human tendency to prefer activity over non-activity is a widely observed phenomenon, even in situations in which inactivity is payoff-maximizing. Such active participation bias, which leads to deviations from payoff-maximizing choices, could stem from different drivers in different situations, including social preferences, social image concerns, (false) beliefs, risk preferences, or simply errors in decision-making. Using a set of experiments, we show that a large majority of participants display an innate preference for activity, even when such confounding factors are ruled out. We also show that active participation bias potentially has huge negative consequences, especially in interaction with narrow framing which can lead individuals to even move further away from the optimal outcome.

• Voting under Salience Bias and Strategic Extremism | with Günnur Ege Bilgin
  Abstract: In the last decade, we have seen populist leaders in many countries such as the USA, Hungary, and Brazil following extreme policies. While most theories suggest that moving to the center from the extremist policies would increase the vote share, moderation does not occur. We believe that limited attention can explain the recent extremism. We created a model of voters with limited attention and the only equilibrium is extremism in at least one policy. We also try to show some empirical evidence such that voters focus more on extreme policies and overstate the importance of those issues.

• Experimental Analysis of Misattribution Models | [draft available upon request]
  Abstract: Attribution Bias (or Misattribution) is shown effective in many economic contexts. However, there is no consensus on the underlying mechanism. I show that existing theories overlap for single period experiences but they differ in multi-period case and at the limit. Hence, I tested the theories via a lab experiment with repeated interactions. The results show that references are a crucial channel in attribution bias. This is the first experiment that tests attribution bias in a multi-period setting.