Good afternoon, ladies and gentlemen, Thank you for having me here. As I have only a single minute to present my dissertation, I have to speak fast. Therefore, I am making more speech errors than usual. Because of the inertia of my articulatory organs, but also, and this is the central idea of my dissertation, because my brain is willing to make errors in order to speed up the mental computation. The starting point of my thesis was Optimality Theory, which I used as a model of linguistic competence. I developed a computational algorithm that aims at finding the best candidate of the candidate set. Simulated Annealing, borrowed from physics, does exactly what our brain does: if it runs slowly, most often it returns the best candidate. But if it runs quickly, the chance increases that it will return some other candidate, and so it will make mistakes. Importantly, it does not return just any candidate. It returns local optima. That is, it explains why certain errors are produced, and why not other errors. I successfully employed this algorithm to simulate Dutch fast speech and other phenomena.