The concept of individual choice is a fundamental aspect when explaining and anticipating behavioural interactions with, and responses to, static and dynamic travel conditions in public transport (PT) systems. However, the empirical grounding of existing models used for forecasting travel demand, which itself is a result of a multitude of individual choices, is often insufficient in terms of detail and accuracy. This thesis explores three aspects, or themes, of PT trips – waiting times, general door-to-door path preferences, with a special emphasis on access and egress trip legs, and service reliability – in order to increase knowledge about how PT passengers interact with PT systems. Using detailed spatiotemporal empirical data from a dedicated survey app and PT fare card transactions, possible cross-sectional relationships between travel conditions and waiting times are analysed, where degrees of mental effort are gauged by an information acquisition proxy. Preferences for complete door-todoor paths are examined by estimation of full path choice models. Finally, longitudinal effects of changing service reliability are analysed using a biennial panel data approach. The constituent studies conclude that there are other explanatory factors than headway that explain waiting times on first boarding stops of PT trips and that possession of knowledge of exact departure times reduces mean waiting times. However, this information factor is not evident in full path choice, where time and effort-related preferences dominate with a consistent individual preference factor. Finally, a statistically significant on-average adaption to changing service reliability is individual-specific and non-symmetrical depending on the direction of reliability change, where a relatively large portion of the affected individuals do not appear to respond to small-scale perturbations of reliability while others do, all other things being equal.