The proposed research project contributes to understanding and implementing familiarity in location-based systems, adding significantly to our knowledge of how to conceptualize familiarity of different environmental features (e.g. landmarks, routes, regions) and providing ways to assess it in-situ based on behavioural data. The project, therefore, has three aims: to provide theoretical, methodological, and applied advancements. Objective 1 is to disentangle and interrelate the different conceptualizations and measurements of familiarity, first, by means of a systematic review. Based on these insights, second, an online study is used to examine different conceptualizations and degrees of familiarity. These results lay the ground for the design of a within-subjects design in-situ study (Objective 2) to collect behavioural correlates of familiarity with different types of features. Three different sensors will be explored as ways to behaviourally assess familiarity during in-situ travel and spatial learning: Mobile eye tracking, high precision GNSS positioning, and head/body-worn Inertial Measurement Units (e.g. acceleration, orientation, etc.), using a subset of Objective 1 subjects and ground-truthed with data collected in Objective 1. Devising an empirical setup suitable to collect behavioural correlates based on these sensors is, therefore, a second major contribution. As a third step (Objective 3), machine learning and deep learning experiments on the behavioural data, singly and in combination, will be used to classify different levels of familiarity reflected in participants activities. These results can contribute to the European Commission's policy "Smart Cities - Smart Living", which interplays with SC4 "Smart, green and integrated transport" of the H2020 work programme: Smart cities can adapt, e.g., public displays to a wayfinder's needs based on their current state of spatial cognition, for which familiarity is a key example.