My research interests are fairly broad; that is probably not such a sensible research strategy if you look at it objectively, but I can't help it. In no particular order:

Models in the social sciences: as a general interest, I can mention the use of formal models (game theoretical or otherwise), on topics that - at first - seem hard to approach using formal methods.

Simulation in the social sciences: whenever it is too complicated to come up with a formal model, it is a useful idea to simulate a model of the behavior of interest, and try to get a better grasp of the essential mechanisms that way.

Trust and cooperation: I am intrigued by the reasons for trustful and trustworhty behavior. Experiments are a nice way to get a handle on this issue, and I ran and still run experiments on "Trust Games" and variants thereof. In addition I try to test the implications of the theories of behavior as tested in the labe, in real world settings such as trust between firms.

Online behavior: another topic that has my attention is the way in which interaction in virtual worlds (on the internet) affects the conditions under which people interact, and how people deal with these different conditions. Do reputation systems in online auctions work? Why are some online interactions working well and others not?

Purchasing and supply, organizational sociology: though at first purchasing and supply was a mere test case for the predictions of my theoretical and empirical work, I slowly grew fond of the topic itself. Something similar goes for organization sociology.

Innovation and its diffusion through networks: how do the connections between people or firms affect their behavior? Under which conditions can we get innovation going?

Decision making and choice under uncertainty: people are notoriously weird when it comes to decision making under circumstances that involve probabilities. Luckily, there is some consistency in the way in which people are weird.

Experts, expertise, and decision support systems: the scientific literature has shown quite conclusively that computer models outperform human experts in a large number of tasks, even tasks that you would swear are the typical domain of human experts. How can that be? And, perhaps even more surprising, computer models are not used that much for prediction purposes. Why is that?