The objective of the postdoctoral thesis is the multilevel analysis of driving behaviour with focus on distraction based on a driving simulator experiment. More specifically, the present postdoctoral thesis is an extension of the PhD thesis of the researcher in which a large driving simulator experiment took place in which 95 participants were asked to drive under different types of distraction (no distraction, conversation with passenger, cell phone use) in different road (urban/rural) and traffic conditions (high/low). Then, a new microscopic database was developed where the average value of all driving performance measures was estimated for a time period of 15 seconds before and 15 seconds after each unexpected incident. To achieve research objectives an advanced statistical analysis methodology was developed which consisted of the development of generalised linear mixed models as well as Structural Equation Models (SEMs). Results indicate that the parameters affecting driving behaviour are different from those that affect the change in behavior that occurs after an unexpected event. This finding is quantified for the first time and can be a be a compass for further investigation of the changes that take place after an unexpected event in driving performance.